Convolutional Neural Networks¶
Project: Write an Algorithm for a Dog Identification App¶
In this notebook, some template code has already been provided for you, and you will need to implement additional functionality to successfully complete this project. You will not need to modify the included code beyond what is requested. Sections that begin with '(IMPLEMENTATION)' in the header indicate that the following block of code will require additional functionality which you must provide. Instructions will be provided for each section, and the specifics of the implementation are marked in the code block with a 'TODO' statement. Please be sure to read the instructions carefully!
Note: Once you have completed all of the code implementations, you need to finalize your work by exporting the iPython Notebook as an HTML document. Before exporting the notebook to html, all of the code cells need to have been run so that reviewers can see the final implementation and output. You can then export the notebook by using the menu above and navigating to \n", "File -> Download as -> HTML (.html). Include the finished document along with this notebook as your submission.
In addition to implementing code, there will be questions that you must answer which relate to the project and your implementation. Each section where you will answer a question is preceded by a 'Question X' header. Carefully read each question and provide thorough answers in the following text boxes that begin with 'Answer:'. Your project submission will be evaluated based on your answers to each of the questions and the implementation you provide.
Note: Code and Markdown cells can be executed using the Shift + Enter keyboard shortcut. Markdown cells can be edited by double-clicking the cell to enter edit mode.
The rubric contains optional "Stand Out Suggestions" for enhancing the project beyond the minimum requirements. If you decide to pursue the "Stand Out Suggestions", you should include the code in this IPython notebook.
Why We're Here¶
In this notebook, you will make the first steps towards developing an algorithm that could be used as part of a mobile or web app. At the end of this project, your code will accept any user-supplied image as input. If a dog is detected in the image, it will provide an estimate of the dog's breed. If a human is detected, it will provide an estimate of the dog breed that is most resembling. The image below displays potential sample output of your finished project (... but we expect that each student's algorithm will behave differently!).
In this real-world setting, you will need to piece together a series of models to perform different tasks; for instance, the algorithm that detects humans in an image will be different from the CNN that infers dog breed. There are many points of possible failure, and no perfect algorithm exists. Your imperfect solution will nonetheless create a fun user experience!
The Road Ahead¶
We break the notebook into separate steps. Feel free to use the links below to navigate the notebook.
- Step 0: Import Datasets
- Step 1: Detect Humans
- Step 2: Detect Dogs
- Step 3: Create a CNN to Classify Dog Breeds (from Scratch)
- Step 4: Use a CNN to Classify Dog Breeds (using Transfer Learning)
- Step 5: Create a CNN to Classify Dog Breeds (using Transfer Learning)
- Step 6: Write your Algorithm
- Step 7: Test Your Algorithm
Step 0: Import Datasets¶
Import Dog Dataset¶
In the code cell below, we import a dataset of dog images. We populate a few variables through the use of the load_files function from the scikit-learn library:
train_files,valid_files,test_files- numpy arrays containing file paths to imagestrain_targets,valid_targets,test_targets- numpy arrays containing onehot-encoded classification labelsdog_names- list of string-valued dog breed names for translating labels
In [4]:
from sklearn.datasets import load_files
from keras.utils import np_utils
import numpy as np
from glob import glob
# define function to load train, test, and validation datasets
def load_dataset(path):
data = load_files(path)
dog_files = np.array(data['filenames'])
dog_targets = np_utils.to_categorical(np.array(data['target']), 133)
return dog_files, dog_targets
# load train, test, and validation datasets
train_files, train_targets = load_dataset('/data/dog_images/train')
valid_files, valid_targets = load_dataset('/data/dog_images/valid')
test_files, test_targets = load_dataset('/data/dog_images/test')
# load list of dog names
dog_names = [item[20:-1].split('.')[1] for item in sorted(glob("/data/dog_images/train/*/"))]
# print statistics about the dataset
print('There are %d total dog categories.' % len(dog_names))
print('There are %s total dog images.\n' % len(np.hstack([train_files, valid_files, test_files])))
print('There are %d training dog images.' % len(train_files))
print('There are %d validation dog images.' % len(valid_files))
print('There are %d test dog images.'% len(test_files))
Using TensorFlow backend.
There are 133 total dog categories. There are 8351 total dog images. There are 6680 training dog images. There are 835 validation dog images. There are 836 test dog images.
Import Human Dataset¶
In the code cell below, we import a dataset of human images, where the file paths are stored in the numpy array human_files.
In [5]:
import random
random.seed(8675309)
# load filenames in shuffled human dataset
human_files = np.array(glob("/data/lfw/*/*"))
random.shuffle(human_files)
# print statistics about the dataset
print('There are %d total human images.' % len(human_files))
There are 13233 total human images.
Step 1: Detect Humans¶
We use OpenCV's implementation of Haar feature-based cascade classifiers to detect human faces in images. OpenCV provides many pre-trained face detectors, stored as XML files on github. We have downloaded one of these detectors and stored it in the haarcascades directory.
In the next code cell, we demonstrate how to use this detector to find human faces in a sample image.
In [6]:
import cv2
import matplotlib.pyplot as plt
%matplotlib inline
def bounding_onhumanface(imgpath):
# extract pre-trained face detector
face_cascade = cv2.CascadeClassifier('haarcascades/haarcascade_frontalface_alt.xml')
# load color (BGR) image
img = cv2.imread(imgpath)
# convert BGR image to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
# find faces in image
faces = face_cascade.detectMultiScale(gray)
# get bounding box for each detected face
for (x,y,w,h) in faces:
# add bounding box to color image
cv2.rectangle(img,(x,y),(x+w,y+h),(255,0,0),2)
# convert BGR image to RGB for plotting
cv_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
# display the image, along with bounding box
plt.imshow(cv_rgb)
plt.show()
# print number of faces detected in the image
print('Number of faces detected:', len(faces))
print("*************************")
bounding_onhumanface(human_files[5])
Number of faces detected: 1 *************************
Before using any of the face detectors, it is standard procedure to convert the images to grayscale. The detectMultiScale function executes the classifier stored in face_cascade and takes the grayscale image as a parameter.
In the above code, faces is a numpy array of detected faces, where each row corresponds to a detected face. Each detected face is a 1D array with four entries that specifies the bounding box of the detected face. The first two entries in the array (extracted in the above code as x and y) specify the horizontal and vertical positions of the top left corner of the bounding box. The last two entries in the array (extracted here as w and h) specify the width and height of the box.
Write a Human Face Detector¶
We can use this procedure to write a function that returns True if a human face is detected in an image and False otherwise. This function, aptly named face_detector, takes a string-valued file path to an image as input and appears in the code block below.
In [7]:
# returns "True" if face is detected in image stored at img_path
def humanface_detector(img_path):
face_cascade = cv2.CascadeClassifier('haarcascades/haarcascade_frontalface_alt.xml')
img = cv2.imread(img_path)
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray)
return len(faces) > 0
(IMPLEMENTATION) Assess the Human Face Detector¶
Question 1: Use the code cell below to test the performance of the face_detector function.
- What percentage of the first 100 images in
human_fileshave a detected human face? - What percentage of the first 100 images in
dog_fileshave a detected human face?
Ideally, we would like 100% of human images with a detected face and 0% of dog images with a detected face. You will see that our algorithm falls short of this goal, but still gives acceptable performance. We extract the file paths for the first 100 images from each of the datasets and store them in the numpy arrays human_files_short and dog_files_short.
Answer:
What percentage of the first 100 images in
human_fileshave a detected human face?
The percentage of human faces detected in humanimg are= 100.000%
What percentage of the first 100 images in
dog_fileshave a detected human face?
The percentage of human faces detected in dogimg are= 11.000%
In [5]:
human_files_short = human_files[:100]
dog_files_short = train_files[:100]
# Do NOT modify the code above this line.
import cv2
print("Testing the performance of the humanface_detector function on the images in human_files_short")
count_humanfaceDetectAlgo_detect_humanface_in_humanimgs=0
for humanimg in human_files_short:
if(humanface_detector(humanimg)):
count_humanfaceDetectAlgo_detect_humanface_in_humanimgs += 1
percentage_human_face_detected_in_humanimg=((count_humanfaceDetectAlgo_detect_humanface_in_humanimgs)/len(human_files_short))*100
print("The percentage of human faces detected in humanimg are={0:3.3f}%".format(percentage_human_face_detected_in_humanimg))
print("Testing the performance of the humanface_detector function on the images in dog_files_short")
count_humanfaceDetectAlgo_detect_humanface_in_dogimgs=0
for dogimg in dog_files_short:
if(humanface_detector(dogimg)):
count_humanfaceDetectAlgo_detect_humanface_in_dogimgs += 1
percentage_humanface_detected_in_dogimg=((count_humanfaceDetectAlgo_detect_humanface_in_dogimgs)/len(dog_files_short))*100
print("The percentage of human faces detected in dogimg are={0:3.3f}% ".format(percentage_humanface_detected_in_dogimg))
Testing the performance of the humanface_detector function on the images in human_files_short The percentage of human faces detected in humanimg are=100.000% Testing the performance of the humanface_detector function on the images in dog_files_short The percentage of human faces detected in dogimg are=11.000%
Question 2: This algorithmic choice necessitates that we communicate to the user that we accept human images only when they provide a clear view of a face (otherwise, we risk having unneccessarily frustrated users!). In your opinion, is this a reasonable expectation to pose on the user? If not, can you think of a way to detect humans in images that does not necessitate an image with a clearly presented face?
Answer:
Since no algorithm can give us 100% sureity that it only built to detect human faces.
In all human face detector algo there is atleast 5% probabiltiy that a non-human face is predicted as human face resulting in False positive and a human face sometimes not predicted due to not clearly presented huamn face in image resulting in True negative. So, i consider it is necessary to tell user beforehand that this algo is for human face detection based on facial feature,shape and size of human face.
One more thing to say to user is that the supplied human face image should be straight as above implemented Haar feature-based cascade classifiers to detect human faces in images is sensitive to rotation varience due the Standard Haar-like features are not rotated to identify rotated human faces.(look below images of girl's face as straight and rotated ).
Haar-like features have been used successfully in image sensors for face tracking and classification problems (Lai et al., 2001; Jones and Viola, 2003; Barreto et al., 2004; Huang and Lai, 2004), however other problems such as hand tracking (Barczak et al., 2005; Micilotta and Bowden, 2004; Kölsch and Turk, 2004) have not been so successful. The main reason for this is the fact that Haar-like features are not invariant over rotation. This means that any object that rotates and is sensitive to angle changes (such as hands) will be difficult to solve using standard Haar-like features. The features that define faces tend to be insensitive to small angle variations and Haar-like features have been used to detect head rotations of as much as 15o from the vertical (Jones and Viola, 2003). When people are standing their head is naturally aligned vertically with respect to gravity and so this rotational sensitivity tends not to be a significant problem for faces. Other body parts such as hands, arms and legs are not normally alligned with the horizontal or vertical axes so are difficult to model with traditional Haar-like features. Researchers have tended to use edge detection or colour based tracking of these parts (Messom et al., 2007). Several researchers have studied the impact of in plane rotations for image sensors with the use of twisted Haar-like feature (45o ) (Lienhart and Maydt, 2002; Lienhart et al., 2003a; 2003b) or diagonal features (Viola and Jones, 2001b) fairly good performance has been achieved. These techniques will have little benefit for problems that are sensitive to rotations, such as hand identification (Barczak et al., 2005; Kölsch and Turk, 2004; Antón-CanalÃs et al., 2005; Stenger et al, 2004; Wachs et al., 2005) which are not aligned to fixed angles (0o , 45o , 90o etc).
In [9]:
#load test files from dog-project/hooman_images
hooman1= "Hooman_images/hooman_straight.jpg"
hooman2= "Hooman_images/hooman_rotated.jpg"
def check(img_path):
if(humanface_detector(img_path)):
bounding_onhumanface(img_path)
print("Hello human ur Face is detected")
else:
print("*****sorry******")
img = cv2.imread(img_path)
cv_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
plt.imshow(cv_rgb)
plt.show()
print("Face not detected")
check(hooman1)
Number of faces detected: 1 ************************* Hello human ur Face is detected
In [10]:
check(hooman2)
*****sorry******
Face not detected
Step 2: Detect Dogs¶
In this section, we use a pre-trained ResNet-50 model to detect dogs in images. Our first line of code downloads the ResNet-50 model, along with weights that have been trained on ImageNet, a very large, very popular dataset used for image classification and other vision tasks. ImageNet contains over 10 million URLs, each linking to an image containing an object from one of 1000 categories. Given an image, this pre-trained ResNet-50 model returns a prediction (derived from the available categories in ImageNet) for the object that is contained in the image.
In [8]:
from keras.applications.resnet50 import ResNet50
# define ResNet50 model
ResNet50_model = ResNet50(weights='imagenet')
Downloading data from https://github.com/fchollet/deep-learning-models/releases/download/v0.2/resnet50_weights_tf_dim_ordering_tf_kernels.h5 102858752/102853048 [==============================] - 3s 0us/step
Pre-process the Data¶
When using TensorFlow as backend, Keras CNNs require a 4D array (which we'll also refer to as a 4D tensor) as input, with shape
$$ (\text{nb_samples}, \text{rows}, \text{columns}, \text{channels}), $$
where nb_samples corresponds to the total number of images (or samples), and rows, columns, and channels correspond to the number of rows, columns, and channels for each image, respectively.
The path_to_tensor function below takes a string-valued file path to a color image as input and returns a 4D tensor suitable for supplying to a Keras CNN. The function first loads the image and resizes it to a square image that is $224 \times 224$ pixels. Next, the image is converted to an array, which is then resized to a 4D tensor. In this case, since we are working with color images, each image has three channels. Likewise, since we are processing a single image (or sample), the returned tensor will always have shape
$$ (1, 224, 224, 3). $$
The paths_to_tensor function takes a numpy array of string-valued image paths as input and returns a 4D tensor with shape
$$ (\text{nb_samples}, 224, 224, 3). $$
Here, nb_samples is the number of samples, or number of images, in the supplied array of image paths. It is best to think of nb_samples as the number of 3D tensors (where each 3D tensor corresponds to a different image) in your dataset!
In [9]:
from keras.preprocessing import image
from tqdm import tqdm
def path_to_tensor(img_path):
# loads RGB image as PIL.Image.Image type
img = image.load_img(img_path, target_size=(224, 224))
# convert PIL.Image.Image type to 3D tensor with shape (224, 224, 3)
x = image.img_to_array(img)
# convert 3D tensor to 4D tensor with shape (1, 224, 224, 3) and return 4D tensor
return np.expand_dims(x, axis=0)
def paths_to_tensor(img_paths):
list_of_tensors = [path_to_tensor(img_path) for img_path in tqdm(img_paths)]
return np.vstack(list_of_tensors)
Making Predictions with ResNet-50¶
Getting the 4D tensor ready for ResNet-50, and for any other pre-trained model in Keras, requires some additional processing. First, the RGB image is converted to BGR by reordering the channels. All pre-trained models have the additional normalization step that the mean pixel (expressed in RGB as $[103.939, 116.779, 123.68]$ and calculated from all pixels in all images in ImageNet) must be subtracted from every pixel in each image. This is implemented in the imported function preprocess_input. If you're curious, you can check the code for preprocess_input here.
Now that we have a way to format our image for supplying to ResNet-50, we are now ready to use the model to extract the predictions. This is accomplished with the predict method, which returns an array whose $i$-th entry is the model's predicted probability that the image belongs to the $i$-th ImageNet category. This is implemented in the ResNet50_predict_labels function below.
By taking the argmax of the predicted probability vector, we obtain an integer corresponding to the model's predicted object class, which we can identify with an object category through the use of this dictionary.
In [10]:
from keras.applications.resnet50 import preprocess_input, decode_predictions
def ResNet50_predict_labels(img_path):
# returns prediction vector for image located at img_path
img = preprocess_input(path_to_tensor(img_path))
return np.argmax(ResNet50_model.predict(img))
In [14]:
img = preprocess_input(path_to_tensor(dog_files_short[0]))
img.shape
Out[14]:
(1, 224, 224, 3)
Write a Dog Detector¶
While looking at the dictionary, you will notice that the categories corresponding to dogs appear in an uninterrupted sequence and correspond to dictionary keys 151-268, inclusive, to include all categories from 'Chihuahua' to 'Mexican hairless'. Thus, in order to check to see if an image is predicted to contain a dog by the pre-trained ResNet-50 model, we need only check if the ResNet50_predict_labels function above returns a value between 151 and 268 (inclusive).
We use these ideas to complete the dog_detector function below, which returns True if a dog is detected in an image (and False if not).
In [11]:
### returns "True" if a dog is detected in the image stored at img_path
def dog_detector(img_path):
prediction = ResNet50_predict_labels(img_path)
return ((prediction <= 268) & (prediction >= 151))
(IMPLEMENTATION) Assess the Dog Detector¶
Question 3: Use the code cell below to test the performance of your dog_detector function.
- What percentage of the images in
human_files_shorthave a detected dog? - What percentage of the images in
dog_files_shorthave a detected dog?
Answer:
What percentage of the images in
human_files_shorthave a detected dog?
The percentage of dog faces detected in humanimg are= 0.000%
What percentage of the images in
dog_files_shorthave a detected dog?
The percentage of dog faces detected in dogimg are= 100.000%
In [10]:
### TODO: Test the performance of the dog_detector function
### on the images in human_files_short and dog_files_short.
print("Testing the performance of the dog_detector function on the images in human_files_short")
count_dogdetectalgo_detect_dogface_in_humanimgs=0
for humanimg in human_files_short:
if(dog_detector(humanimg)):
count_dogdetectalgo_detect_dogface_in_humanimgs += 1
percentage_dog_face_detected_in_humanimg=((count_dogdetectalgo_detect_dogface_in_humanimgs)/len(human_files_short))*100
print("The percentage of dog faces detected in humanimg are={0:3.3f}%".format(percentage_dog_face_detected_in_humanimg))
print("Testing the performance of the dog_detector function on the images in dog_files_short")
count_dogdetectalgo_detect_dogface_in_dogimgs=0
for dogimg in dog_files_short:
if(dog_detector(dogimg)):
count_dogdetectalgo_detect_dogface_in_dogimgs += 1
percentage_dog_face_detected_in_dogimg=((count_dogdetectalgo_detect_dogface_in_dogimgs)/len(dog_files_short))*100
print("The percentage of dog faces detected in dogimg are={0:3.3f}% ".format(percentage_dog_face_detected_in_dogimg))
Testing the performance of the dog_detector function on the images in human_files_short The percentage of dog faces detected in humanimg are=0.000% Testing the performance of the dog_detector function on the images in dog_files_short The percentage of dog faces detected in dogimg are=100.000%
Step 3: Create a CNN to Classify Dog Breeds (from Scratch)¶
Now that we have functions for detecting humans and dogs in images, we need a way to predict breed from images. In this step, you will create a CNN that classifies dog breeds. You must create your CNN from scratch (so, you can't use transfer learning yet!), and you must attain a test accuracy of at least 1%. In Step 5 of this notebook, you will have the opportunity to use transfer learning to create a CNN that attains greatly improved accuracy.
Be careful with adding too many trainable layers! More parameters means longer training, which means you are more likely to need a GPU to accelerate the training process. Thankfully, Keras provides a handy estimate of the time that each epoch is likely to take; you can extrapolate this estimate to figure out how long it will take for your algorithm to train.
We mention that the task of assigning breed to dogs from images is considered exceptionally challenging. To see why, consider that even a human would have great difficulty in distinguishing between a Brittany and a Welsh Springer Spaniel.
| Brittany | Welsh Springer Spaniel |
|---|---|
 |
 |
It is not difficult to find other dog breed pairs with minimal inter-class variation (for instance, Curly-Coated Retrievers and American Water Spaniels).
| Curly-Coated Retriever | American Water Spaniel |
|---|---|
 |
 |
Likewise, recall that labradors come in yellow, chocolate, and black. Your vision-based algorithm will have to conquer this high intra-class variation to determine how to classify all of these different shades as the same breed.
| Yellow Labrador | Chocolate Labrador | Black Labrador |
|---|---|---|
 |
 |
 |
We also mention that random chance presents an exceptionally low bar: setting aside the fact that the classes are slightly imabalanced, a random guess will provide a correct answer roughly 1 in 133 times, which corresponds to an accuracy of less than 1%.
Remember that the practice is far ahead of the theory in deep learning. Experiment with many different architectures, and trust your intuition. And, of course, have fun!
Pre-process the Data¶
We rescale the images by dividing every pixel in every image by 255.
In [12]:
from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True
# pre-process the data for Keras
train_tensors = paths_to_tensor(train_files).astype('float32')/255
valid_tensors = paths_to_tensor(valid_files).astype('float32')/255
test_tensors = paths_to_tensor(test_files).astype('float32')/255
100%|██████████| 6680/6680 [01:15<00:00, 88.99it/s] 100%|██████████| 835/835 [00:08<00:00, 100.67it/s] 100%|██████████| 836/836 [00:08<00:00, 100.99it/s]
(IMPLEMENTATION) Model Architecture¶
Create a CNN to classify dog breed. At the end of your code cell block, summarize the layers of your model by executing the line:
model.summary()
We have imported some Python modules to get you started, but feel free to import as many modules as you need. If you end up getting stuck, here's a hint that specifies a model that trains relatively fast on CPU and attains >1% test accuracy in 5 epochs:
Question 4: Outline the steps you took to get to your final CNN architecture and your reasoning at each step. If you chose to use the hinted architecture above, describe why you think that CNN architecture should work well for the image classification task.
Answer: I choose to use above hinted arch. as series of convolutional layer with increasing filter size will allow the network to identify the increasingly complex pattern to better ditnisguish between different dog breed. The pooling layer is usefull to reduce the dimensionalty of convolutional layer it was getting and also reduce overfitting and increase validation accuracy.
In [13]:
from keras.layers import Conv2D, MaxPooling2D, GlobalAveragePooling2D
from keras.layers import Dropout, Flatten, Dense
from keras.models import Sequential
model = Sequential()
### TODO: Define your architecture.
#train_tensors.shape=(6680, 224, 224, 3)
model.add(Conv2D(filters=16, kernel_size=2, padding='same', activation='relu',
input_shape=(224, 224, 3)))
model.add(MaxPooling2D(pool_size=2))
model.add(Conv2D(filters=32, kernel_size=2, padding='same', activation='relu'))
model.add(MaxPooling2D(pool_size=2))
model.add(Conv2D(filters=64, kernel_size=2, padding='same', activation='relu'))
model.add(MaxPooling2D(pool_size=2))
model.add(GlobalAveragePooling2D())
model.add(Dense(len(dog_names),activation="softmax"))
model.summary()
_________________________________________________________________ Layer (type) Output Shape Param # ================================================================= conv2d_1 (Conv2D) (None, 224, 224, 16) 208 _________________________________________________________________ max_pooling2d_2 (MaxPooling2 (None, 112, 112, 16) 0 _________________________________________________________________ conv2d_2 (Conv2D) (None, 112, 112, 32) 2080 _________________________________________________________________ max_pooling2d_3 (MaxPooling2 (None, 56, 56, 32) 0 _________________________________________________________________ conv2d_3 (Conv2D) (None, 56, 56, 64) 8256 _________________________________________________________________ max_pooling2d_4 (MaxPooling2 (None, 28, 28, 64) 0 _________________________________________________________________ global_average_pooling2d_1 ( (None, 64) 0 _________________________________________________________________ dense_1 (Dense) (None, 133) 8645 ================================================================= Total params: 19,189 Trainable params: 19,189 Non-trainable params: 0 _________________________________________________________________
Compile the Model¶
In [14]:
model.compile(optimizer='rmsprop', loss='categorical_crossentropy', metrics=['accuracy'])
(IMPLEMENTATION) Train the Model¶
Train your model in the code cell below. Use model checkpointing to save the model that attains the best validation loss.
You are welcome to augment the training data, but this is not a requirement.
In [18]:
from keras.callbacks import ModelCheckpoint
### TODO: specify the number of epochs that you would like to use to train the model.
epochs = 10
### Do NOT modify the code below this line.
checkpointer = ModelCheckpoint(filepath='saved_models2/weights.best.from_scratch.hdf5',
verbose=1, save_best_only=True)
model.fit(train_tensors, train_targets,
validation_data=(valid_tensors, valid_targets),
epochs=epochs, batch_size=20, callbacks=[checkpointer], verbose=1)
Train on 6680 samples, validate on 835 samples Epoch 1/10 6660/6680 [============================>.] - ETA: 0s - loss: 4.8835 - acc: 0.0081Epoch 00001: val_loss improved from inf to 4.87083, saving model to saved_models2/weights.best.from_scratch.hdf5 6680/6680 [==============================] - 22s 3ms/step - loss: 4.8835 - acc: 0.0081 - val_loss: 4.8708 - val_acc: 0.0108 Epoch 2/10 6660/6680 [============================>.] - ETA: 0s - loss: 4.8690 - acc: 0.0086Epoch 00002: val_loss improved from 4.87083 to 4.86213, saving model to saved_models2/weights.best.from_scratch.hdf5 6680/6680 [==============================] - 21s 3ms/step - loss: 4.8691 - acc: 0.0085 - val_loss: 4.8621 - val_acc: 0.0120 Epoch 3/10 6660/6680 [============================>.] - ETA: 0s - loss: 4.8468 - acc: 0.0144Epoch 00003: val_loss improved from 4.86213 to 4.83081, saving model to saved_models2/weights.best.from_scratch.hdf5 6680/6680 [==============================] - 21s 3ms/step - loss: 4.8470 - acc: 0.0144 - val_loss: 4.8308 - val_acc: 0.0144 Epoch 4/10 6660/6680 [============================>.] - ETA: 0s - loss: 4.7923 - acc: 0.0215Epoch 00004: val_loss improved from 4.83081 to 4.78709, saving model to saved_models2/weights.best.from_scratch.hdf5 6680/6680 [==============================] - 21s 3ms/step - loss: 4.7928 - acc: 0.0214 - val_loss: 4.7871 - val_acc: 0.0216 Epoch 5/10 6660/6680 [============================>.] - ETA: 0s - loss: 4.7523 - acc: 0.0204Epoch 00005: val_loss improved from 4.78709 to 4.76332, saving model to saved_models2/weights.best.from_scratch.hdf5 6680/6680 [==============================] - 21s 3ms/step - loss: 4.7527 - acc: 0.0204 - val_loss: 4.7633 - val_acc: 0.0216 Epoch 6/10 6660/6680 [============================>.] - ETA: 0s - loss: 4.7251 - acc: 0.0269Epoch 00006: val_loss improved from 4.76332 to 4.75332, saving model to saved_models2/weights.best.from_scratch.hdf5 6680/6680 [==============================] - 21s 3ms/step - loss: 4.7252 - acc: 0.0268 - val_loss: 4.7533 - val_acc: 0.0251 Epoch 7/10 6660/6680 [============================>.] - ETA: 0s - loss: 4.6976 - acc: 0.0272Epoch 00007: val_loss improved from 4.75332 to 4.72692, saving model to saved_models2/weights.best.from_scratch.hdf5 6680/6680 [==============================] - 21s 3ms/step - loss: 4.6984 - acc: 0.0271 - val_loss: 4.7269 - val_acc: 0.0335 Epoch 8/10 6660/6680 [============================>.] - ETA: 0s - loss: 4.6712 - acc: 0.0327Epoch 00008: val_loss improved from 4.72692 to 4.72575, saving model to saved_models2/weights.best.from_scratch.hdf5 6680/6680 [==============================] - 21s 3ms/step - loss: 4.6704 - acc: 0.0328 - val_loss: 4.7258 - val_acc: 0.0275 Epoch 9/10 6660/6680 [============================>.] - ETA: 0s - loss: 4.6415 - acc: 0.0348Epoch 00009: val_loss improved from 4.72575 to 4.71126, saving model to saved_models2/weights.best.from_scratch.hdf5 6680/6680 [==============================] - 21s 3ms/step - loss: 4.6417 - acc: 0.0349 - val_loss: 4.7113 - val_acc: 0.0251 Epoch 10/10 6660/6680 [============================>.] - ETA: 0s - loss: 4.6147 - acc: 0.0351Epoch 00010: val_loss improved from 4.71126 to 4.68738, saving model to saved_models2/weights.best.from_scratch.hdf5 6680/6680 [==============================] - 21s 3ms/step - loss: 4.6143 - acc: 0.0350 - val_loss: 4.6874 - val_acc: 0.0335
Out[18]:
<keras.callbacks.History at 0x7f8d62f97518>
Load the Model with the Best Validation Loss¶
In [14]:
model.load_weights('saved_models2/weights.best.from_scratch.hdf5')
Test the Model¶
Try out your model on the test dataset of dog images. Ensure that your test accuracy is greater than 1%.
In [15]:
# get index of predicted dog breed for each image in test set
dog_breed_predictions = [np.argmax(model.predict(np.expand_dims(tensor, axis=0))) for tensor in test_tensors]
# report test accuracy
test_accuracy = 100*np.sum(np.array(dog_breed_predictions)==np.argmax(test_targets, axis=1))/len(dog_breed_predictions)
#https://youtu.be/3sDYifgjFck?t=56 follow this
print('Test accuracy: %.4f%%' % test_accuracy)
Test accuracy: 4.1866%
In [16]:
train_tensors.shape
Out[16]:
(6680, 224, 224, 3)
In [15]:
bottleneck_features = np.load('/data/bottleneck_features/DogVGG16Data.npz')
train_VGG16 = bottleneck_features['train']
valid_VGG16 = bottleneck_features['valid']
test_VGG16 = bottleneck_features['test']
In [30]:
train_VGG16.shape #this parameters and weights will be the input to new layer for training data..
Out[30]:
(6680, 7, 7, 512)
Model Architecture¶
The model uses the the pre-trained VGG-16 model as a fixed feature extractor, where the last convolutional output of VGG-16 is fed as input to our model. We only add a global average pooling layer and a fully connected layer, where the latter contains one node for each dog category and is equipped with a softmax.
In [16]:
VGG16_model = Sequential()
VGG16_model.add(GlobalAveragePooling2D(input_shape=train_VGG16.shape[1:]))
VGG16_model.add(Dense(133, activation='softmax'))
VGG16_model.summary()
_________________________________________________________________ Layer (type) Output Shape Param # ================================================================= global_average_pooling2d_2 ( (None, 512) 0 _________________________________________________________________ dense_2 (Dense) (None, 133) 68229 ================================================================= Total params: 68,229 Trainable params: 68,229 Non-trainable params: 0 _________________________________________________________________
Compile the Model¶
In [17]:
VGG16_model.compile(loss='categorical_crossentropy', optimizer='rmsprop', metrics=['accuracy'])
Train the Model¶
In [20]:
### TODO: Train the model.
from keras.callbacks import ModelCheckpoint
batch_size = [20,35,37,40]
epochs = [20,35,40,50]
fitingdict_vgg16={'Batch_Size':[],
'Epochs':[],
'Test_Accuracy':[]}
for bs in batch_size:
for ep in epochs:
checkpointer = ModelCheckpoint(filepath='saved_models1/weights.best.vgg16_bs'+str(bs)+'_ep'+str(ep)+'.hdf5',
verbose=1, save_best_only=True)
print("\nBatch size={0} Epoch={1}".format(bs,ep))
VGG16_model.fit(train_VGG16, train_targets,validation_data=(valid_VGG16, valid_targets),
epochs=ep , batch_size=bs,
callbacks=[checkpointer],verbose=1)
#LOAD the model with Best validation loss
VGG16_model.load_weights('saved_models1/weights.best.vgg16_bs'+str(bs)+'_ep'+str(ep)+'.hdf5')
VGG16_predictions = [np.argmax(VGG16_model.predict(np.expand_dims(feature, axis=0))) for feature in test_VGG16]
test_accuracy = 100*np.sum(np.array(VGG16_predictions)==np.argmax(test_targets, axis=1))/len(VGG16_predictions)
fitingdict_vgg16['Batch_Size'].append(bs)
fitingdict_vgg16['Epochs'].append(ep)
fitingdict_vgg16['Test_Accuracy'].append(test_accuracy)
Batch size=20 Epoch=20 Train on 6680 samples, validate on 835 samples Epoch 1/20 6640/6680 [============================>.] - ETA: 0s - loss: 12.2849 - acc: 0.1268Epoch 00001: val_loss improved from inf to 10.73861, saving model to saved_models1/weights.best.vgg16_bs20_ep20.hdf5 6680/6680 [==============================] - 2s 316us/step - loss: 12.2704 - acc: 0.1277 - val_loss: 10.7386 - val_acc: 0.2024 Epoch 2/20 6660/6680 [============================>.] - ETA: 0s - loss: 9.7665 - acc: 0.2959Epoch 00002: val_loss improved from 10.73861 to 9.56120, saving model to saved_models1/weights.best.vgg16_bs20_ep20.hdf5 6680/6680 [==============================] - 2s 257us/step - loss: 9.7663 - acc: 0.2963 - val_loss: 9.5612 - val_acc: 0.2982 Epoch 3/20 6520/6680 [============================>.] - ETA: 0s - loss: 9.0206 - acc: 0.3758Epoch 00003: val_loss improved from 9.56120 to 9.32497, saving model to saved_models1/weights.best.vgg16_bs20_ep20.hdf5 6680/6680 [==============================] - 2s 252us/step - loss: 9.0173 - acc: 0.3759 - val_loss: 9.3250 - val_acc: 0.3257 Epoch 4/20 6480/6680 [============================>.] - ETA: 0s - loss: 8.7739 - acc: 0.4105Epoch 00004: val_loss improved from 9.32497 to 9.16428, saving model to saved_models1/weights.best.vgg16_bs20_ep20.hdf5 6680/6680 [==============================] - 2s 252us/step - loss: 8.7702 - acc: 0.4103 - val_loss: 9.1643 - val_acc: 0.3473 Epoch 5/20 6460/6680 [============================>.] - ETA: 0s - loss: 8.4450 - acc: 0.4353Epoch 00005: val_loss improved from 9.16428 to 8.77212, saving model to saved_models1/weights.best.vgg16_bs20_ep20.hdf5 6680/6680 [==============================] - 2s 252us/step - loss: 8.4474 - acc: 0.4346 - val_loss: 8.7721 - val_acc: 0.3737 Epoch 6/20 6500/6680 [============================>.] - ETA: 0s - loss: 8.1363 - acc: 0.4585Epoch 00006: val_loss improved from 8.77212 to 8.64753, saving model to saved_models1/weights.best.vgg16_bs20_ep20.hdf5 6680/6680 [==============================] - 2s 252us/step - loss: 8.1527 - acc: 0.4575 - val_loss: 8.6475 - val_acc: 0.3808 Epoch 7/20 6460/6680 [============================>.] - ETA: 0s - loss: 8.0196 - acc: 0.4746Epoch 00007: val_loss improved from 8.64753 to 8.58411, saving model to saved_models1/weights.best.vgg16_bs20_ep20.hdf5 6680/6680 [==============================] - 2s 254us/step - loss: 8.0126 - acc: 0.4750 - val_loss: 8.5841 - val_acc: 0.3832 Epoch 8/20 6660/6680 [============================>.] - ETA: 0s - loss: 7.9130 - acc: 0.4859Epoch 00008: val_loss improved from 8.58411 to 8.48738, saving model to saved_models1/weights.best.vgg16_bs20_ep20.hdf5 6680/6680 [==============================] - 2s 254us/step - loss: 7.9183 - acc: 0.4856 - val_loss: 8.4874 - val_acc: 0.3916 Epoch 9/20 6660/6680 [============================>.] - ETA: 0s - loss: 7.8093 - acc: 0.4935Epoch 00009: val_loss improved from 8.48738 to 8.48704, saving model to saved_models1/weights.best.vgg16_bs20_ep20.hdf5 6680/6680 [==============================] - 2s 252us/step - loss: 7.8138 - acc: 0.4931 - val_loss: 8.4870 - val_acc: 0.3940 Epoch 10/20 6640/6680 [============================>.] - ETA: 0s - loss: 7.7342 - acc: 0.5005Epoch 00010: val_loss improved from 8.48704 to 8.42266, saving model to saved_models1/weights.best.vgg16_bs20_ep20.hdf5 6680/6680 [==============================] - 2s 254us/step - loss: 7.7386 - acc: 0.5003 - val_loss: 8.4227 - val_acc: 0.3964 Epoch 11/20 6520/6680 [============================>.] - ETA: 0s - loss: 7.6543 - acc: 0.5109Epoch 00011: val_loss improved from 8.42266 to 8.36733, saving model to saved_models1/weights.best.vgg16_bs20_ep20.hdf5 6680/6680 [==============================] - 2s 252us/step - loss: 7.6441 - acc: 0.5112 - val_loss: 8.3673 - val_acc: 0.4048 Epoch 12/20 6620/6680 [============================>.] - ETA: 0s - loss: 7.6121 - acc: 0.5181Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 2s 252us/step - loss: 7.6091 - acc: 0.5181 - val_loss: 8.3853 - val_acc: 0.4072 Epoch 13/20 6500/6680 [============================>.] - ETA: 0s - loss: 7.5441 - acc: 0.5202Epoch 00013: val_loss improved from 8.36733 to 8.28634, saving model to saved_models1/weights.best.vgg16_bs20_ep20.hdf5 6680/6680 [==============================] - 2s 252us/step - loss: 7.5376 - acc: 0.5208 - val_loss: 8.2863 - val_acc: 0.4096 Epoch 14/20 6460/6680 [============================>.] - ETA: 0s - loss: 7.4309 - acc: 0.5300Epoch 00014: val_loss improved from 8.28634 to 8.14191, saving model to saved_models1/weights.best.vgg16_bs20_ep20.hdf5 6680/6680 [==============================] - 2s 253us/step - loss: 7.4346 - acc: 0.5296 - val_loss: 8.1419 - val_acc: 0.4240 Epoch 15/20 6520/6680 [============================>.] - ETA: 0s - loss: 7.3041 - acc: 0.5314Epoch 00015: val_loss improved from 8.14191 to 8.03198, saving model to saved_models1/weights.best.vgg16_bs20_ep20.hdf5 6680/6680 [==============================] - 2s 252us/step - loss: 7.2951 - acc: 0.5316 - val_loss: 8.0320 - val_acc: 0.4156 Epoch 16/20 6640/6680 [============================>.] - ETA: 0s - loss: 7.1039 - acc: 0.5453Epoch 00016: val_loss improved from 8.03198 to 7.92229, saving model to saved_models1/weights.best.vgg16_bs20_ep20.hdf5 6680/6680 [==============================] - 2s 255us/step - loss: 7.1048 - acc: 0.5454 - val_loss: 7.9223 - val_acc: 0.4144 Epoch 17/20 6500/6680 [============================>.] - ETA: 0s - loss: 7.0511 - acc: 0.5526Epoch 00017: val_loss improved from 7.92229 to 7.80997, saving model to saved_models1/weights.best.vgg16_bs20_ep20.hdf5 6680/6680 [==============================] - 2s 252us/step - loss: 7.0452 - acc: 0.5530 - val_loss: 7.8100 - val_acc: 0.4383 Epoch 18/20 6640/6680 [============================>.] - ETA: 0s - loss: 6.9734 - acc: 0.5557Epoch 00018: val_loss improved from 7.80997 to 7.73699, saving model to saved_models1/weights.best.vgg16_bs20_ep20.hdf5 6680/6680 [==============================] - 2s 254us/step - loss: 6.9709 - acc: 0.5558 - val_loss: 7.7370 - val_acc: 0.4299 Epoch 19/20 6640/6680 [============================>.] - ETA: 0s - loss: 6.8487 - acc: 0.5640Epoch 00019: val_loss improved from 7.73699 to 7.67507, saving model to saved_models1/weights.best.vgg16_bs20_ep20.hdf5 6680/6680 [==============================] - 2s 257us/step - loss: 6.8464 - acc: 0.5642 - val_loss: 7.6751 - val_acc: 0.4503 Epoch 20/20 6640/6680 [============================>.] - ETA: 0s - loss: 6.7425 - acc: 0.5745Epoch 00020: val_loss improved from 7.67507 to 7.61237, saving model to saved_models1/weights.best.vgg16_bs20_ep20.hdf5 6680/6680 [==============================] - 2s 253us/step - loss: 6.7553 - acc: 0.5738 - val_loss: 7.6124 - val_acc: 0.4587 Batch size=20 Epoch=35 Train on 6680 samples, validate on 835 samples Epoch 1/35 6480/6680 [============================>.] - ETA: 0s - loss: 6.7527 - acc: 0.5753Epoch 00001: val_loss improved from inf to 7.63061, saving model to saved_models1/weights.best.vgg16_bs20_ep35.hdf5 6680/6680 [==============================] - 2s 252us/step - loss: 6.7377 - acc: 0.5762 - val_loss: 7.6306 - val_acc: 0.4515 Epoch 2/35 6480/6680 [============================>.] - ETA: 0s - loss: 6.7121 - acc: 0.5755Epoch 00002: val_loss improved from 7.63061 to 7.56560, saving model to saved_models1/weights.best.vgg16_bs20_ep35.hdf5 6680/6680 [==============================] - 2s 252us/step - loss: 6.6910 - acc: 0.5763 - val_loss: 7.5656 - val_acc: 0.4527 Epoch 3/35 6500/6680 [============================>.] - ETA: 0s - loss: 6.5811 - acc: 0.5822Epoch 00003: val_loss improved from 7.56560 to 7.44745, saving model to saved_models1/weights.best.vgg16_bs20_ep35.hdf5 6680/6680 [==============================] - 2s 252us/step - loss: 6.5799 - acc: 0.5825 - val_loss: 7.4475 - val_acc: 0.4611 Epoch 4/35 6660/6680 [============================>.] - ETA: 0s - loss: 6.5353 - acc: 0.5889Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 6.5304 - acc: 0.5891 - val_loss: 7.4680 - val_acc: 0.4695 Epoch 5/35 6520/6680 [============================>.] - ETA: 0s - loss: 6.4319 - acc: 0.5896Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 2s 249us/step - loss: 6.4369 - acc: 0.5889 - val_loss: 7.5604 - val_acc: 0.4539 Epoch 6/35 6480/6680 [============================>.] - ETA: 0s - loss: 6.3749 - acc: 0.5968Epoch 00006: val_loss improved from 7.44745 to 7.36995, saving model to saved_models1/weights.best.vgg16_bs20_ep35.hdf5 6680/6680 [==============================] - 2s 252us/step - loss: 6.3789 - acc: 0.5967 - val_loss: 7.3700 - val_acc: 0.4671 Epoch 7/35 6580/6680 [============================>.] - ETA: 0s - loss: 6.3626 - acc: 0.5983Epoch 00007: val_loss improved from 7.36995 to 7.35530, saving model to saved_models1/weights.best.vgg16_bs20_ep35.hdf5 6680/6680 [==============================] - 2s 250us/step - loss: 6.3519 - acc: 0.5991 - val_loss: 7.3553 - val_acc: 0.4659 Epoch 8/35 6480/6680 [============================>.] - ETA: 0s - loss: 6.3235 - acc: 0.6020Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 6.3257 - acc: 0.6019 - val_loss: 7.3729 - val_acc: 0.4707 Epoch 9/35 6580/6680 [============================>.] - ETA: 0s - loss: 6.2815 - acc: 0.6015Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 2s 249us/step - loss: 6.2895 - acc: 0.6009 - val_loss: 7.3566 - val_acc: 0.4719 Epoch 10/35 6580/6680 [============================>.] - ETA: 0s - loss: 6.1421 - acc: 0.6081Epoch 00010: val_loss improved from 7.35530 to 7.30314, saving model to saved_models1/weights.best.vgg16_bs20_ep35.hdf5 6680/6680 [==============================] - 2s 251us/step - loss: 6.1543 - acc: 0.6072 - val_loss: 7.3031 - val_acc: 0.4659 Epoch 11/35 6560/6680 [============================>.] - ETA: 0s - loss: 6.0993 - acc: 0.6142Epoch 00011: val_loss improved from 7.30314 to 7.24965, saving model to saved_models1/weights.best.vgg16_bs20_ep35.hdf5 6680/6680 [==============================] - 2s 250us/step - loss: 6.1104 - acc: 0.6136 - val_loss: 7.2497 - val_acc: 0.4754 Epoch 12/35 6480/6680 [============================>.] - ETA: 0s - loss: 6.0601 - acc: 0.6151Epoch 00012: val_loss improved from 7.24965 to 7.11197, saving model to saved_models1/weights.best.vgg16_bs20_ep35.hdf5 6680/6680 [==============================] - 2s 252us/step - loss: 6.0437 - acc: 0.6162 - val_loss: 7.1120 - val_acc: 0.4814 Epoch 13/35 6600/6680 [============================>.] - ETA: 0s - loss: 5.9624 - acc: 0.6227Epoch 00013: val_loss improved from 7.11197 to 7.06896, saving model to saved_models1/weights.best.vgg16_bs20_ep35.hdf5 6680/6680 [==============================] - 2s 249us/step - loss: 5.9803 - acc: 0.6217 - val_loss: 7.0690 - val_acc: 0.4838 Epoch 14/35 6460/6680 [============================>.] - ETA: 0s - loss: 5.9562 - acc: 0.6246Epoch 00014: val_loss improved from 7.06896 to 7.02231, saving model to saved_models1/weights.best.vgg16_bs20_ep35.hdf5 6680/6680 [==============================] - 2s 251us/step - loss: 5.9641 - acc: 0.6241 - val_loss: 7.0223 - val_acc: 0.4886 Epoch 15/35 6500/6680 [============================>.] - ETA: 0s - loss: 5.9696 - acc: 0.6249Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.9518 - acc: 0.6257 - val_loss: 7.0988 - val_acc: 0.4874 Epoch 16/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.9491 - acc: 0.6262Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.9418 - acc: 0.6268 - val_loss: 7.0677 - val_acc: 0.4934 Epoch 17/35 6500/6680 [============================>.] - ETA: 0s - loss: 5.9558 - acc: 0.6262Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.9401 - acc: 0.6272 - val_loss: 7.0414 - val_acc: 0.4934 Epoch 18/35 6460/6680 [============================>.] - ETA: 0s - loss: 5.9518 - acc: 0.6280Epoch 00018: val_loss improved from 7.02231 to 7.01481, saving model to saved_models1/weights.best.vgg16_bs20_ep35.hdf5 6680/6680 [==============================] - 2s 252us/step - loss: 5.9391 - acc: 0.6289 - val_loss: 7.0148 - val_acc: 0.4958 Epoch 19/35 6580/6680 [============================>.] - ETA: 0s - loss: 5.9467 - acc: 0.6274Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 2s 249us/step - loss: 5.9307 - acc: 0.6283 - val_loss: 7.1441 - val_acc: 0.4934 Epoch 20/35 6460/6680 [============================>.] - ETA: 0s - loss: 5.8875 - acc: 0.6277Epoch 00020: val_loss improved from 7.01481 to 6.99397, saving model to saved_models1/weights.best.vgg16_bs20_ep35.hdf5 6680/6680 [==============================] - 2s 252us/step - loss: 5.8822 - acc: 0.6281 - val_loss: 6.9940 - val_acc: 0.5042 Epoch 21/35 6480/6680 [============================>.] - ETA: 0s - loss: 5.8406 - acc: 0.6338Epoch 00021: val_loss improved from 6.99397 to 6.97883, saving model to saved_models1/weights.best.vgg16_bs20_ep35.hdf5 6680/6680 [==============================] - 2s 252us/step - loss: 5.8324 - acc: 0.6341 - val_loss: 6.9788 - val_acc: 0.4958 Epoch 22/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.8655 - acc: 0.6337Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 2s 249us/step - loss: 5.8296 - acc: 0.6359 - val_loss: 6.9797 - val_acc: 0.4982 Epoch 23/35 6540/6680 [============================>.] - ETA: 0s - loss: 5.8313 - acc: 0.6343Epoch 00023: val_loss improved from 6.97883 to 6.95965, saving model to saved_models1/weights.best.vgg16_bs20_ep35.hdf5 6680/6680 [==============================] - 2s 251us/step - loss: 5.8284 - acc: 0.6343 - val_loss: 6.9596 - val_acc: 0.5018 Epoch 24/35 6640/6680 [============================>.] - ETA: 0s - loss: 5.8228 - acc: 0.6360Epoch 00024: val_loss improved from 6.95965 to 6.94867, saving model to saved_models1/weights.best.vgg16_bs20_ep35.hdf5 6680/6680 [==============================] - 2s 254us/step - loss: 5.8197 - acc: 0.6361 - val_loss: 6.9487 - val_acc: 0.4970 Epoch 25/35 6460/6680 [============================>.] - ETA: 0s - loss: 5.8139 - acc: 0.6370Epoch 00025: val_loss improved from 6.94867 to 6.92395, saving model to saved_models1/weights.best.vgg16_bs20_ep35.hdf5 6680/6680 [==============================] - 2s 252us/step - loss: 5.8155 - acc: 0.6370 - val_loss: 6.9239 - val_acc: 0.5054 Epoch 26/35 6580/6680 [============================>.] - ETA: 0s - loss: 5.7984 - acc: 0.6383Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.8154 - acc: 0.6373 - val_loss: 6.9685 - val_acc: 0.4946 Epoch 27/35 6620/6680 [============================>.] - ETA: 0s - loss: 5.8188 - acc: 0.6376Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 2s 248us/step - loss: 5.8100 - acc: 0.6382 - val_loss: 7.0006 - val_acc: 0.4946 Epoch 28/35 6540/6680 [============================>.] - ETA: 0s - loss: 5.8034 - acc: 0.6387Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.8097 - acc: 0.6383 - val_loss: 7.0077 - val_acc: 0.4982 Epoch 29/35 6660/6680 [============================>.] - ETA: 0s - loss: 5.8084 - acc: 0.6380Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.8127 - acc: 0.6377 - val_loss: 7.0003 - val_acc: 0.4874 Epoch 30/35 6500/6680 [============================>.] - ETA: 0s - loss: 5.7806 - acc: 0.6400Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.8083 - acc: 0.6383 - val_loss: 7.0214 - val_acc: 0.4934 Epoch 31/35 6500/6680 [============================>.] - ETA: 0s - loss: 5.7889 - acc: 0.6398Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.8090 - acc: 0.6386 - val_loss: 7.0100 - val_acc: 0.4922 Epoch 32/35 6560/6680 [============================>.] - ETA: 0s - loss: 5.7916 - acc: 0.6390Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 2s 249us/step - loss: 5.8107 - acc: 0.6379 - val_loss: 6.9448 - val_acc: 0.5090 Epoch 33/35 6460/6680 [============================>.] - ETA: 0s - loss: 5.8106 - acc: 0.6385Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.8050 - acc: 0.6389 - val_loss: 7.0753 - val_acc: 0.4994 Epoch 34/35 6640/6680 [============================>.] - ETA: 0s - loss: 5.7994 - acc: 0.6396Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 2s 252us/step - loss: 5.8062 - acc: 0.6391 - val_loss: 7.0301 - val_acc: 0.4982 Epoch 35/35 6660/6680 [============================>.] - ETA: 0s - loss: 5.8049 - acc: 0.6392Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 2s 252us/step - loss: 5.8093 - acc: 0.6389 - val_loss: 7.0263 - val_acc: 0.5102 Batch size=20 Epoch=40 Train on 6680 samples, validate on 835 samples Epoch 1/40 6660/6680 [============================>.] - ETA: 0s - loss: 5.8101 - acc: 0.6378Epoch 00001: val_loss improved from inf to 6.97348, saving model to saved_models1/weights.best.vgg16_bs20_ep40.hdf5 6680/6680 [==============================] - 2s 253us/step - loss: 5.8120 - acc: 0.6377 - val_loss: 6.9735 - val_acc: 0.5018 Epoch 2/40 6460/6680 [============================>.] - ETA: 0s - loss: 5.8191 - acc: 0.6373Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.8112 - acc: 0.6377 - val_loss: 6.9982 - val_acc: 0.5018 Epoch 3/40 6460/6680 [============================>.] - ETA: 0s - loss: 5.8087 - acc: 0.6379Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.8128 - acc: 0.6377 - val_loss: 7.0417 - val_acc: 0.4874 Epoch 4/40 6460/6680 [============================>.] - ETA: 0s - loss: 5.7907 - acc: 0.6393Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.8105 - acc: 0.6380 - val_loss: 7.0377 - val_acc: 0.4946 Epoch 5/40 6480/6680 [============================>.] - ETA: 0s - loss: 5.8067 - acc: 0.6386Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.8090 - acc: 0.6385 - val_loss: 7.0683 - val_acc: 0.4862 Epoch 6/40 6480/6680 [============================>.] - ETA: 0s - loss: 5.8255 - acc: 0.6372Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.8127 - acc: 0.6380 - val_loss: 6.9991 - val_acc: 0.4970 Epoch 7/40 6480/6680 [============================>.] - ETA: 0s - loss: 5.7969 - acc: 0.6397Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.8094 - acc: 0.6386 - val_loss: 6.9921 - val_acc: 0.5042 Epoch 8/40 6540/6680 [============================>.] - ETA: 0s - loss: 5.8161 - acc: 0.6381Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.8125 - acc: 0.6383 - val_loss: 7.0207 - val_acc: 0.4982 Epoch 9/40 6640/6680 [============================>.] - ETA: 0s - loss: 5.8081 - acc: 0.6389Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 2s 252us/step - loss: 5.8095 - acc: 0.6388 - val_loss: 7.0493 - val_acc: 0.4910 Epoch 10/40 6600/6680 [============================>.] - ETA: 0s - loss: 5.8204 - acc: 0.6380Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 2s 248us/step - loss: 5.8086 - acc: 0.6388 - val_loss: 7.0098 - val_acc: 0.4994 Epoch 11/40 6640/6680 [============================>.] - ETA: 0s - loss: 5.8114 - acc: 0.6380Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 2s 252us/step - loss: 5.8080 - acc: 0.6382 - val_loss: 7.0471 - val_acc: 0.4946 Epoch 12/40 6660/6680 [============================>.] - ETA: 0s - loss: 5.8131 - acc: 0.6384Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 2s 252us/step - loss: 5.8053 - acc: 0.6389 - val_loss: 7.0390 - val_acc: 0.5030 Epoch 13/40 6600/6680 [============================>.] - ETA: 0s - loss: 5.8001 - acc: 0.6392Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.8078 - acc: 0.6388 - val_loss: 7.0246 - val_acc: 0.5006 Epoch 14/40 6460/6680 [============================>.] - ETA: 0s - loss: 5.8071 - acc: 0.6392Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 2s 252us/step - loss: 5.8089 - acc: 0.6391 - val_loss: 7.0758 - val_acc: 0.4982 Epoch 15/40 6480/6680 [============================>.] - ETA: 0s - loss: 5.7673 - acc: 0.6415Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.8046 - acc: 0.6392 - val_loss: 7.0242 - val_acc: 0.5006 Epoch 16/40 6480/6680 [============================>.] - ETA: 0s - loss: 5.8293 - acc: 0.6375Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.8043 - acc: 0.6391 - val_loss: 7.0295 - val_acc: 0.5066 Epoch 17/40 6640/6680 [============================>.] - ETA: 0s - loss: 5.8217 - acc: 0.6380Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 2s 252us/step - loss: 5.8085 - acc: 0.6388 - val_loss: 7.0546 - val_acc: 0.4886 Epoch 18/40 6500/6680 [============================>.] - ETA: 0s - loss: 5.8020 - acc: 0.6392Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.8024 - acc: 0.6392 - val_loss: 7.0540 - val_acc: 0.4982 Epoch 19/40 6500/6680 [============================>.] - ETA: 0s - loss: 5.7943 - acc: 0.6378Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.7758 - acc: 0.6391 - val_loss: 7.0433 - val_acc: 0.4970 Epoch 20/40 6540/6680 [============================>.] - ETA: 0s - loss: 5.7275 - acc: 0.6425Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.7162 - acc: 0.6431 - val_loss: 6.9876 - val_acc: 0.5078 Epoch 21/40 6660/6680 [============================>.] - ETA: 0s - loss: 5.7091 - acc: 0.6449Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.7113 - acc: 0.6448 - val_loss: 7.0182 - val_acc: 0.4982 Epoch 22/40 6500/6680 [============================>.] - ETA: 0s - loss: 5.7361 - acc: 0.6428Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.7094 - acc: 0.6445 - val_loss: 7.0028 - val_acc: 0.5018 Epoch 23/40 6500/6680 [============================>.] - ETA: 0s - loss: 5.6955 - acc: 0.6460Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.7061 - acc: 0.6454 - val_loss: 6.9924 - val_acc: 0.5042 Epoch 24/40 6460/6680 [============================>.] - ETA: 0s - loss: 5.7130 - acc: 0.6447Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.7058 - acc: 0.6452 - val_loss: 7.0113 - val_acc: 0.5030 Epoch 25/40 6500/6680 [============================>.] - ETA: 0s - loss: 5.6970 - acc: 0.6460Epoch 00025: val_loss improved from 6.97348 to 6.94948, saving model to saved_models1/weights.best.vgg16_bs20_ep40.hdf5 6680/6680 [==============================] - 2s 252us/step - loss: 5.7051 - acc: 0.6455 - val_loss: 6.9495 - val_acc: 0.5042 Epoch 26/40 6480/6680 [============================>.] - ETA: 0s - loss: 5.6991 - acc: 0.6461Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.7046 - acc: 0.6458 - val_loss: 6.9543 - val_acc: 0.5042 Epoch 27/40 6460/6680 [============================>.] - ETA: 0s - loss: 5.7026 - acc: 0.6454Epoch 00027: val_loss improved from 6.94948 to 6.93504, saving model to saved_models1/weights.best.vgg16_bs20_ep40.hdf5 6680/6680 [==============================] - 2s 253us/step - loss: 5.7054 - acc: 0.6452 - val_loss: 6.9350 - val_acc: 0.5054 Epoch 28/40 6620/6680 [============================>.] - ETA: 0s - loss: 5.7167 - acc: 0.6444Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 2s 252us/step - loss: 5.7064 - acc: 0.6451 - val_loss: 6.9522 - val_acc: 0.5078 Epoch 29/40 6500/6680 [============================>.] - ETA: 0s - loss: 5.6977 - acc: 0.6455Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.7082 - acc: 0.6449 - val_loss: 6.9490 - val_acc: 0.4994 Epoch 30/40 6480/6680 [============================>.] - ETA: 0s - loss: 5.6980 - acc: 0.6463Epoch 00030: val_loss improved from 6.93504 to 6.91137, saving model to saved_models1/weights.best.vgg16_bs20_ep40.hdf5 6680/6680 [==============================] - 2s 252us/step - loss: 5.7011 - acc: 0.6461 - val_loss: 6.9114 - val_acc: 0.5054 Epoch 31/40 6540/6680 [============================>.] - ETA: 0s - loss: 5.7165 - acc: 0.6448Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.7053 - acc: 0.6455 - val_loss: 6.9656 - val_acc: 0.5042 Epoch 32/40 6460/6680 [============================>.] - ETA: 0s - loss: 5.7016 - acc: 0.6457Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 2s 254us/step - loss: 5.7020 - acc: 0.6457 - val_loss: 6.9697 - val_acc: 0.5042 Epoch 33/40 6580/6680 [============================>.] - ETA: 0s - loss: 5.6943 - acc: 0.6459Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 2s 262us/step - loss: 5.7032 - acc: 0.6454 - val_loss: 6.9606 - val_acc: 0.4970 Epoch 34/40 6600/6680 [============================>.] - ETA: 0s - loss: 5.6944 - acc: 0.6459Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 2s 261us/step - loss: 5.7010 - acc: 0.6455 - val_loss: 6.9941 - val_acc: 0.5030 Epoch 35/40 6600/6680 [============================>.] - ETA: 0s - loss: 5.6861 - acc: 0.6452Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 2s 262us/step - loss: 5.6928 - acc: 0.6448 - val_loss: 7.0507 - val_acc: 0.5006 Epoch 36/40 6600/6680 [============================>.] - ETA: 0s - loss: 5.6705 - acc: 0.6438Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 2s 262us/step - loss: 5.6678 - acc: 0.6440 - val_loss: 6.9767 - val_acc: 0.5006 Epoch 37/40 6620/6680 [============================>.] - ETA: 0s - loss: 5.6428 - acc: 0.6473Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 2s 260us/step - loss: 5.6452 - acc: 0.6472 - val_loss: 6.9230 - val_acc: 0.5030 Epoch 38/40 6600/6680 [============================>.] - ETA: 0s - loss: 5.6485 - acc: 0.6471Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 2s 259us/step - loss: 5.6316 - acc: 0.6481 - val_loss: 6.9211 - val_acc: 0.5090 Epoch 39/40 6620/6680 [============================>.] - ETA: 0s - loss: 5.6294 - acc: 0.6486Epoch 00039: val_loss improved from 6.91137 to 6.85904, saving model to saved_models1/weights.best.vgg16_bs20_ep40.hdf5 6680/6680 [==============================] - 2s 264us/step - loss: 5.6238 - acc: 0.6490 - val_loss: 6.8590 - val_acc: 0.5066 Epoch 40/40 6600/6680 [============================>.] - ETA: 0s - loss: 5.6215 - acc: 0.6491Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 2s 264us/step - loss: 5.6193 - acc: 0.6493 - val_loss: 6.8874 - val_acc: 0.5030 Batch size=20 Epoch=50 Train on 6680 samples, validate on 835 samples Epoch 1/50 6500/6680 [============================>.] - ETA: 0s - loss: 5.6236 - acc: 0.6494Epoch 00001: val_loss improved from inf to 6.86228, saving model to saved_models1/weights.best.vgg16_bs20_ep50.hdf5 6680/6680 [==============================] - 2s 251us/step - loss: 5.6193 - acc: 0.6497 - val_loss: 6.8623 - val_acc: 0.5090 Epoch 2/50 6540/6680 [============================>.] - ETA: 0s - loss: 5.6337 - acc: 0.6486Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 2s 249us/step - loss: 5.6127 - acc: 0.6499 - val_loss: 6.8927 - val_acc: 0.5078 Epoch 3/50 6480/6680 [============================>.] - ETA: 0s - loss: 5.5925 - acc: 0.6519Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.6092 - acc: 0.6506 - val_loss: 6.9195 - val_acc: 0.4970 Epoch 4/50 6500/6680 [============================>.] - ETA: 0s - loss: 5.6022 - acc: 0.6508Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.6129 - acc: 0.6501 - val_loss: 6.8660 - val_acc: 0.4982 Epoch 5/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6085 - acc: 0.6506Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.6069 - acc: 0.6507 - val_loss: 6.8889 - val_acc: 0.5042 Epoch 6/50 6480/6680 [============================>.] - ETA: 0s - loss: 5.6087 - acc: 0.6512Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.6049 - acc: 0.6515 - val_loss: 6.9809 - val_acc: 0.4970 Epoch 7/50 6460/6680 [============================>.] - ETA: 0s - loss: 5.6143 - acc: 0.6506Epoch 00007: val_loss improved from 6.86228 to 6.85760, saving model to saved_models1/weights.best.vgg16_bs20_ep50.hdf5 6680/6680 [==============================] - 2s 251us/step - loss: 5.6070 - acc: 0.6510 - val_loss: 6.8576 - val_acc: 0.5090 Epoch 8/50 6540/6680 [============================>.] - ETA: 0s - loss: 5.6167 - acc: 0.6509Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.6027 - acc: 0.6518 - val_loss: 6.9685 - val_acc: 0.5006 Epoch 9/50 6540/6680 [============================>.] - ETA: 0s - loss: 5.6001 - acc: 0.6517Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.6058 - acc: 0.6513 - val_loss: 6.8841 - val_acc: 0.5018 Epoch 10/50 6620/6680 [============================>.] - ETA: 0s - loss: 5.5968 - acc: 0.6521Epoch 00010: val_loss improved from 6.85760 to 6.85536, saving model to saved_models1/weights.best.vgg16_bs20_ep50.hdf5 6680/6680 [==============================] - 2s 258us/step - loss: 5.6045 - acc: 0.6516 - val_loss: 6.8554 - val_acc: 0.5042 Epoch 11/50 6640/6680 [============================>.] - ETA: 0s - loss: 5.6054 - acc: 0.6517Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 2s 254us/step - loss: 5.6032 - acc: 0.6518 - val_loss: 6.8731 - val_acc: 0.5114 Epoch 12/50 6460/6680 [============================>.] - ETA: 0s - loss: 5.5978 - acc: 0.6520Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.6041 - acc: 0.6516 - val_loss: 6.9051 - val_acc: 0.5018 Epoch 13/50 6660/6680 [============================>.] - ETA: 0s - loss: 5.6087 - acc: 0.6518Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 2s 252us/step - loss: 5.6040 - acc: 0.6521 - val_loss: 6.8800 - val_acc: 0.5030 Epoch 14/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.5864 - acc: 0.6531Epoch 00014: val_loss improved from 6.85536 to 6.84438, saving model to saved_models1/weights.best.vgg16_bs20_ep50.hdf5 6680/6680 [==============================] - 2s 252us/step - loss: 5.6046 - acc: 0.6519 - val_loss: 6.8444 - val_acc: 0.5102 Epoch 15/50 6500/6680 [============================>.] - ETA: 0s - loss: 5.6025 - acc: 0.6517Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.6035 - acc: 0.6516 - val_loss: 6.8538 - val_acc: 0.5078 Epoch 16/50 6440/6680 [===========================>..] - ETA: 0s - loss: 5.6108 - acc: 0.6514Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.6046 - acc: 0.6518 - val_loss: 6.8618 - val_acc: 0.5102 Epoch 17/50 6660/6680 [============================>.] - ETA: 0s - loss: 5.6027 - acc: 0.6520Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 2s 252us/step - loss: 5.6053 - acc: 0.6518 - val_loss: 6.8877 - val_acc: 0.5114 Epoch 18/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6183 - acc: 0.6512Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.6040 - acc: 0.6521 - val_loss: 6.9141 - val_acc: 0.5066 Epoch 19/50 6480/6680 [============================>.] - ETA: 0s - loss: 5.6259 - acc: 0.6503Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 2s 252us/step - loss: 5.6029 - acc: 0.6516 - val_loss: 6.8703 - val_acc: 0.5198 Epoch 20/50 6500/6680 [============================>.] - ETA: 0s - loss: 5.6243 - acc: 0.6503Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.6055 - acc: 0.6515 - val_loss: 6.8752 - val_acc: 0.5162 Epoch 21/50 6480/6680 [============================>.] - ETA: 0s - loss: 5.6237 - acc: 0.6503Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.6037 - acc: 0.6515 - val_loss: 6.8614 - val_acc: 0.5150 Epoch 22/50 6480/6680 [============================>.] - ETA: 0s - loss: 5.6189 - acc: 0.6509Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 2s 249us/step - loss: 5.6027 - acc: 0.6519 - val_loss: 6.8699 - val_acc: 0.5114 Epoch 23/50 6480/6680 [============================>.] - ETA: 0s - loss: 5.5987 - acc: 0.6522Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.6056 - acc: 0.6516 - val_loss: 6.8861 - val_acc: 0.5078 Epoch 24/50 6480/6680 [============================>.] - ETA: 0s - loss: 5.5674 - acc: 0.6543Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.6044 - acc: 0.6519 - val_loss: 6.8636 - val_acc: 0.5162 Epoch 25/50 6460/6680 [============================>.] - ETA: 0s - loss: 5.6322 - acc: 0.6498Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.6035 - acc: 0.6516 - val_loss: 6.9273 - val_acc: 0.5102 Epoch 26/50 6480/6680 [============================>.] - ETA: 0s - loss: 5.6000 - acc: 0.6520Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.6037 - acc: 0.6518 - val_loss: 6.9008 - val_acc: 0.5126 Epoch 27/50 6480/6680 [============================>.] - ETA: 0s - loss: 5.6044 - acc: 0.6517Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 2s 252us/step - loss: 5.6055 - acc: 0.6516 - val_loss: 6.8891 - val_acc: 0.5054 Epoch 28/50 6660/6680 [============================>.] - ETA: 0s - loss: 5.6110 - acc: 0.6517Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.6038 - acc: 0.6521 - val_loss: 6.8867 - val_acc: 0.5138 Epoch 29/50 6480/6680 [============================>.] - ETA: 0s - loss: 5.5969 - acc: 0.6523Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.6055 - acc: 0.6518 - val_loss: 6.9275 - val_acc: 0.5090 Epoch 30/50 6620/6680 [============================>.] - ETA: 0s - loss: 5.6087 - acc: 0.6517Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 2s 252us/step - loss: 5.6049 - acc: 0.6518 - val_loss: 6.9071 - val_acc: 0.5102 Epoch 31/50 6460/6680 [============================>.] - ETA: 0s - loss: 5.6134 - acc: 0.6511Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 2s 252us/step - loss: 5.6047 - acc: 0.6516 - val_loss: 6.8817 - val_acc: 0.5090 Epoch 32/50 6460/6680 [============================>.] - ETA: 0s - loss: 5.6100 - acc: 0.6512Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.6024 - acc: 0.6516 - val_loss: 6.8849 - val_acc: 0.5186 Epoch 33/50 6640/6680 [============================>.] - ETA: 0s - loss: 5.6003 - acc: 0.6523Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.6030 - acc: 0.6521 - val_loss: 6.8701 - val_acc: 0.5114 Epoch 34/50 6660/6680 [============================>.] - ETA: 0s - loss: 5.5992 - acc: 0.6523Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.6042 - acc: 0.6519 - val_loss: 6.9030 - val_acc: 0.5102 Epoch 35/50 6460/6680 [============================>.] - ETA: 0s - loss: 5.6342 - acc: 0.6498Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.6031 - acc: 0.6518 - val_loss: 6.8711 - val_acc: 0.5054 Epoch 36/50 6460/6680 [============================>.] - ETA: 0s - loss: 5.5934 - acc: 0.6526Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.6046 - acc: 0.6519 - val_loss: 6.8646 - val_acc: 0.5198 Epoch 37/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.5769 - acc: 0.6535Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.6043 - acc: 0.6518 - val_loss: 6.8826 - val_acc: 0.5078 Epoch 38/50 6660/6680 [============================>.] - ETA: 0s - loss: 5.6001 - acc: 0.6523Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.6003 - acc: 0.6522 - val_loss: 6.8627 - val_acc: 0.5102 Epoch 39/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6200 - acc: 0.6511Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.6063 - acc: 0.6518 - val_loss: 6.8709 - val_acc: 0.5162 Epoch 40/50 6660/6680 [============================>.] - ETA: 0s - loss: 5.6044 - acc: 0.6517Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 2s 252us/step - loss: 5.6046 - acc: 0.6516 - val_loss: 6.8543 - val_acc: 0.5162 Epoch 41/50 6460/6680 [============================>.] - ETA: 0s - loss: 5.6184 - acc: 0.6506Epoch 00041: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.6047 - acc: 0.6515 - val_loss: 6.8579 - val_acc: 0.5126 Epoch 42/50 6640/6680 [============================>.] - ETA: 0s - loss: 5.6153 - acc: 0.6512Epoch 00042: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.6034 - acc: 0.6519 - val_loss: 6.8870 - val_acc: 0.5054 Epoch 43/50 6660/6680 [============================>.] - ETA: 0s - loss: 5.6044 - acc: 0.6521Epoch 00043: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.6045 - acc: 0.6521 - val_loss: 6.8622 - val_acc: 0.5138 Epoch 44/50 6540/6680 [============================>.] - ETA: 0s - loss: 5.6018 - acc: 0.6520Epoch 00044: val_loss did not improve 6680/6680 [==============================] - 2s 250us/step - loss: 5.6051 - acc: 0.6518 - val_loss: 6.8974 - val_acc: 0.5066 Epoch 45/50 6440/6680 [===========================>..] - ETA: 0s - loss: 5.5969 - acc: 0.6525Epoch 00045: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.6034 - acc: 0.6521 - val_loss: 6.8866 - val_acc: 0.5018 Epoch 46/50 6480/6680 [============================>.] - ETA: 0s - loss: 5.5850 - acc: 0.6531Epoch 00046: val_loss did not improve 6680/6680 [==============================] - 2s 253us/step - loss: 5.6036 - acc: 0.6519 - val_loss: 6.8485 - val_acc: 0.5102 Epoch 47/50 6580/6680 [============================>.] - ETA: 0s - loss: 5.6030 - acc: 0.6518Epoch 00047: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.6060 - acc: 0.6516 - val_loss: 6.8473 - val_acc: 0.5138 Epoch 48/50 6640/6680 [============================>.] - ETA: 0s - loss: 5.6035 - acc: 0.6520Epoch 00048: val_loss did not improve 6680/6680 [==============================] - 2s 252us/step - loss: 5.6062 - acc: 0.6518 - val_loss: 6.8772 - val_acc: 0.5114 Epoch 49/50 6500/6680 [============================>.] - ETA: 0s - loss: 5.5894 - acc: 0.6529Epoch 00049: val_loss did not improve 6680/6680 [==============================] - 2s 251us/step - loss: 5.6044 - acc: 0.6519 - val_loss: 6.9241 - val_acc: 0.5138 Epoch 50/50 6640/6680 [============================>.] - ETA: 0s - loss: 5.5952 - acc: 0.6524Epoch 00050: val_loss did not improve 6680/6680 [==============================] - 2s 253us/step - loss: 5.6027 - acc: 0.6519 - val_loss: 6.8733 - val_acc: 0.5090 Batch size=35 Epoch=20 Train on 6680 samples, validate on 835 samples Epoch 1/20 6545/6680 [============================>.] - ETA: 0s - loss: 5.5901 - acc: 0.6526Epoch 00001: val_loss improved from inf to 6.87290, saving model to saved_models1/weights.best.vgg16_bs35_ep20.hdf5 6680/6680 [==============================] - 1s 172us/step - loss: 5.6050 - acc: 0.6516 - val_loss: 6.8729 - val_acc: 0.5078 Epoch 2/20 6370/6680 [===========================>..] - ETA: 0s - loss: 5.5953 - acc: 0.6523Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 1s 172us/step - loss: 5.6035 - acc: 0.6518 - val_loss: 6.8882 - val_acc: 0.5138 Epoch 3/20 6370/6680 [===========================>..] - ETA: 0s - loss: 5.6127 - acc: 0.6512Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 1s 172us/step - loss: 5.6056 - acc: 0.6516 - val_loss: 6.8785 - val_acc: 0.5042 Epoch 4/20 6615/6680 [============================>.] - ETA: 0s - loss: 5.6223 - acc: 0.6508Epoch 00004: val_loss improved from 6.87290 to 6.85918, saving model to saved_models1/weights.best.vgg16_bs35_ep20.hdf5 6680/6680 [==============================] - 1s 172us/step - loss: 5.6038 - acc: 0.6519 - val_loss: 6.8592 - val_acc: 0.5078 Epoch 5/20 6545/6680 [============================>.] - ETA: 0s - loss: 5.5898 - acc: 0.6529Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6071 - acc: 0.6518 - val_loss: 6.8654 - val_acc: 0.5090 Epoch 6/20 6405/6680 [===========================>..] - ETA: 0s - loss: 5.5723 - acc: 0.6539Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 172us/step - loss: 5.6059 - acc: 0.6518 - val_loss: 6.8818 - val_acc: 0.5090 Epoch 7/20 6615/6680 [============================>.] - ETA: 0s - loss: 5.6120 - acc: 0.6512Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6057 - acc: 0.6516 - val_loss: 6.8714 - val_acc: 0.5042 Epoch 8/20 6510/6680 [============================>.] - ETA: 0s - loss: 5.6189 - acc: 0.6508Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6062 - acc: 0.6516 - val_loss: 6.8600 - val_acc: 0.5090 Epoch 9/20 6580/6680 [============================>.] - ETA: 0s - loss: 5.6097 - acc: 0.6514Epoch 00009: val_loss improved from 6.85918 to 6.83809, saving model to saved_models1/weights.best.vgg16_bs35_ep20.hdf5 6680/6680 [==============================] - 1s 172us/step - loss: 5.6029 - acc: 0.6518 - val_loss: 6.8381 - val_acc: 0.5078 Epoch 10/20 6650/6680 [============================>.] - ETA: 0s - loss: 5.6009 - acc: 0.6522Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6047 - acc: 0.6519 - val_loss: 6.9139 - val_acc: 0.5042 Epoch 11/20 6510/6680 [============================>.] - ETA: 0s - loss: 5.6332 - acc: 0.6501Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6049 - acc: 0.6518 - val_loss: 6.8590 - val_acc: 0.5162 Epoch 12/20 6580/6680 [============================>.] - ETA: 0s - loss: 5.6105 - acc: 0.6511Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6062 - acc: 0.6513 - val_loss: 6.8521 - val_acc: 0.5114 Epoch 13/20 6510/6680 [============================>.] - ETA: 0s - loss: 5.6302 - acc: 0.6504Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6051 - acc: 0.6519 - val_loss: 6.8982 - val_acc: 0.5066 Epoch 14/20 6615/6680 [============================>.] - ETA: 0s - loss: 5.6012 - acc: 0.6522Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 169us/step - loss: 5.6046 - acc: 0.6519 - val_loss: 6.8895 - val_acc: 0.5078 Epoch 15/20 6580/6680 [============================>.] - ETA: 0s - loss: 5.5961 - acc: 0.6521Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6041 - acc: 0.6516 - val_loss: 6.8623 - val_acc: 0.5126 Epoch 16/20 6580/6680 [============================>.] - ETA: 0s - loss: 5.6100 - acc: 0.6517Epoch 00016: val_loss improved from 6.83809 to 6.83095, saving model to saved_models1/weights.best.vgg16_bs35_ep20.hdf5 6680/6680 [==============================] - 1s 173us/step - loss: 5.6032 - acc: 0.6521 - val_loss: 6.8310 - val_acc: 0.5114 Epoch 17/20 6580/6680 [============================>.] - ETA: 0s - loss: 5.5967 - acc: 0.6523Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6046 - acc: 0.6518 - val_loss: 6.8367 - val_acc: 0.5114 Epoch 18/20 6615/6680 [============================>.] - ETA: 0s - loss: 5.6159 - acc: 0.6509Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6047 - acc: 0.6516 - val_loss: 6.8710 - val_acc: 0.5066 Epoch 19/20 6650/6680 [============================>.] - ETA: 0s - loss: 5.6063 - acc: 0.6519Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 169us/step - loss: 5.6029 - acc: 0.6521 - val_loss: 6.8551 - val_acc: 0.5102 Epoch 20/20 6440/6680 [===========================>..] - ETA: 0s - loss: 5.6001 - acc: 0.6520Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 172us/step - loss: 5.6040 - acc: 0.6518 - val_loss: 6.8704 - val_acc: 0.5126 Batch size=35 Epoch=35 Train on 6680 samples, validate on 835 samples Epoch 1/35 6545/6680 [============================>.] - ETA: 0s - loss: 5.6030 - acc: 0.6519Epoch 00001: val_loss improved from inf to 6.86478, saving model to saved_models1/weights.best.vgg16_bs35_ep35.hdf5 6680/6680 [==============================] - 1s 172us/step - loss: 5.6032 - acc: 0.6519 - val_loss: 6.8648 - val_acc: 0.5054 Epoch 2/35 6615/6680 [============================>.] - ETA: 0s - loss: 5.5987 - acc: 0.6522Epoch 00002: val_loss improved from 6.86478 to 6.82769, saving model to saved_models1/weights.best.vgg16_bs35_ep35.hdf5 6680/6680 [==============================] - 1s 172us/step - loss: 5.6045 - acc: 0.6518 - val_loss: 6.8277 - val_acc: 0.5162 Epoch 3/35 6650/6680 [============================>.] - ETA: 0s - loss: 5.5999 - acc: 0.6520Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 1s 172us/step - loss: 5.6061 - acc: 0.6516 - val_loss: 6.8400 - val_acc: 0.5042 Epoch 4/35 6440/6680 [===========================>..] - ETA: 0s - loss: 5.6159 - acc: 0.6511Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 1s 172us/step - loss: 5.6048 - acc: 0.6518 - val_loss: 6.8558 - val_acc: 0.5078 Epoch 5/35 6545/6680 [============================>.] - ETA: 0s - loss: 5.6015 - acc: 0.6521Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6041 - acc: 0.6519 - val_loss: 6.8625 - val_acc: 0.5018 Epoch 6/35 6440/6680 [===========================>..] - ETA: 0s - loss: 5.5929 - acc: 0.6526Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 172us/step - loss: 5.6042 - acc: 0.6519 - val_loss: 6.8489 - val_acc: 0.5114 Epoch 7/35 6510/6680 [============================>.] - ETA: 0s - loss: 5.6062 - acc: 0.6518Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6035 - acc: 0.6519 - val_loss: 6.8747 - val_acc: 0.5114 Epoch 8/35 6580/6680 [============================>.] - ETA: 0s - loss: 5.6144 - acc: 0.6511Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6027 - acc: 0.6518 - val_loss: 6.8634 - val_acc: 0.5150 Epoch 9/35 6510/6680 [============================>.] - ETA: 0s - loss: 5.6312 - acc: 0.6502Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6037 - acc: 0.6519 - val_loss: 6.8810 - val_acc: 0.5102 Epoch 10/35 6510/6680 [============================>.] - ETA: 0s - loss: 5.5876 - acc: 0.6528Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 172us/step - loss: 5.6046 - acc: 0.6518 - val_loss: 6.8562 - val_acc: 0.5126 Epoch 11/35 6510/6680 [============================>.] - ETA: 0s - loss: 5.5930 - acc: 0.6525Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6050 - acc: 0.6518 - val_loss: 6.8600 - val_acc: 0.5114 Epoch 12/35 6615/6680 [============================>.] - ETA: 0s - loss: 5.6195 - acc: 0.6511Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6058 - acc: 0.6519 - val_loss: 6.9067 - val_acc: 0.5078 Epoch 13/35 6580/6680 [============================>.] - ETA: 0s - loss: 5.6026 - acc: 0.6518Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6056 - acc: 0.6516 - val_loss: 6.8904 - val_acc: 0.5090 Epoch 14/35 6335/6680 [===========================>..] - ETA: 0s - loss: 5.6106 - acc: 0.6515Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6058 - acc: 0.6516 - val_loss: 6.8788 - val_acc: 0.5114 Epoch 15/35 6545/6680 [============================>.] - ETA: 0s - loss: 5.6226 - acc: 0.6504Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6055 - acc: 0.6515 - val_loss: 6.8786 - val_acc: 0.5054 Epoch 16/35 6545/6680 [============================>.] - ETA: 0s - loss: 5.5861 - acc: 0.6530Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 172us/step - loss: 5.6035 - acc: 0.6519 - val_loss: 6.8733 - val_acc: 0.5090 Epoch 17/35 6545/6680 [============================>.] - ETA: 0s - loss: 5.6031 - acc: 0.6515Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6033 - acc: 0.6515 - val_loss: 6.8490 - val_acc: 0.5114 Epoch 18/35 6580/6680 [============================>.] - ETA: 0s - loss: 5.6089 - acc: 0.6515Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6046 - acc: 0.6518 - val_loss: 6.8579 - val_acc: 0.5114 Epoch 19/35 6510/6680 [============================>.] - ETA: 0s - loss: 5.5955 - acc: 0.6522Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6051 - acc: 0.6516 - val_loss: 6.8765 - val_acc: 0.5042 Epoch 20/35 6405/6680 [===========================>..] - ETA: 0s - loss: 5.5843 - acc: 0.6528Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 172us/step - loss: 5.6053 - acc: 0.6515 - val_loss: 6.8854 - val_acc: 0.5078 Epoch 21/35 6440/6680 [===========================>..] - ETA: 0s - loss: 5.5922 - acc: 0.6526Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6036 - acc: 0.6519 - val_loss: 6.8616 - val_acc: 0.5030 Epoch 22/35 6510/6680 [============================>.] - ETA: 0s - loss: 5.6065 - acc: 0.6518Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 1s 172us/step - loss: 5.6038 - acc: 0.6519 - val_loss: 6.8547 - val_acc: 0.5126 Epoch 23/35 6510/6680 [============================>.] - ETA: 0s - loss: 5.5945 - acc: 0.6524Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6041 - acc: 0.6518 - val_loss: 6.8560 - val_acc: 0.5126 Epoch 24/35 6545/6680 [============================>.] - ETA: 0s - loss: 5.6247 - acc: 0.6507Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6051 - acc: 0.6519 - val_loss: 6.8705 - val_acc: 0.5114 Epoch 25/35 6510/6680 [============================>.] - ETA: 0s - loss: 5.5956 - acc: 0.6524Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 1s 173us/step - loss: 5.6052 - acc: 0.6518 - val_loss: 6.8385 - val_acc: 0.5090 Epoch 26/35 6475/6680 [============================>.] - ETA: 0s - loss: 5.6052 - acc: 0.6517Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 1s 172us/step - loss: 5.6045 - acc: 0.6518 - val_loss: 6.8809 - val_acc: 0.5078 Epoch 27/35 6580/6680 [============================>.] - ETA: 0s - loss: 5.5929 - acc: 0.6526Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6057 - acc: 0.6518 - val_loss: 6.8526 - val_acc: 0.5090 Epoch 28/35 6545/6680 [============================>.] - ETA: 0s - loss: 5.6195 - acc: 0.6510Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6049 - acc: 0.6519 - val_loss: 6.8894 - val_acc: 0.5042 Epoch 29/35 6650/6680 [============================>.] - ETA: 0s - loss: 5.6041 - acc: 0.6520Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6055 - acc: 0.6519 - val_loss: 6.8577 - val_acc: 0.5090 Epoch 30/35 6475/6680 [============================>.] - ETA: 0s - loss: 5.6025 - acc: 0.6519Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6043 - acc: 0.6518 - val_loss: 6.8665 - val_acc: 0.5102 Epoch 31/35 6475/6680 [============================>.] - ETA: 0s - loss: 5.6162 - acc: 0.6511Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6052 - acc: 0.6518 - val_loss: 6.8533 - val_acc: 0.5114 Epoch 32/35 6440/6680 [===========================>..] - ETA: 0s - loss: 5.6124 - acc: 0.6511Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6038 - acc: 0.6516 - val_loss: 6.8764 - val_acc: 0.5102 Epoch 33/35 6510/6680 [============================>.] - ETA: 0s - loss: 5.6315 - acc: 0.6499Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6056 - acc: 0.6515 - val_loss: 6.8899 - val_acc: 0.5066 Epoch 34/35 6580/6680 [============================>.] - ETA: 0s - loss: 5.6080 - acc: 0.6518Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6037 - acc: 0.6521 - val_loss: 6.9018 - val_acc: 0.5054 Epoch 35/35 6650/6680 [============================>.] - ETA: 0s - loss: 5.5985 - acc: 0.6523Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6047 - acc: 0.6519 - val_loss: 6.8915 - val_acc: 0.5042 Batch size=35 Epoch=40 Train on 6680 samples, validate on 835 samples Epoch 1/40 6650/6680 [============================>.] - ETA: 0s - loss: 5.5962 - acc: 0.6523Epoch 00001: val_loss improved from inf to 6.83195, saving model to saved_models1/weights.best.vgg16_bs35_ep40.hdf5 6680/6680 [==============================] - 1s 172us/step - loss: 5.6049 - acc: 0.6518 - val_loss: 6.8320 - val_acc: 0.5126 Epoch 2/40 6510/6680 [============================>.] - ETA: 0s - loss: 5.5891 - acc: 0.6528Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6037 - acc: 0.6519 - val_loss: 6.8679 - val_acc: 0.5114 Epoch 3/40 6335/6680 [===========================>..] - ETA: 0s - loss: 5.6187 - acc: 0.6508Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 1s 169us/step - loss: 5.6043 - acc: 0.6516 - val_loss: 6.8734 - val_acc: 0.5054 Epoch 4/40 6545/6680 [============================>.] - ETA: 0s - loss: 5.6007 - acc: 0.6519Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6057 - acc: 0.6516 - val_loss: 6.8666 - val_acc: 0.5126 Epoch 5/40 6580/6680 [============================>.] - ETA: 0s - loss: 5.5982 - acc: 0.6520Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6037 - acc: 0.6516 - val_loss: 6.8424 - val_acc: 0.5090 Epoch 6/40 6545/6680 [============================>.] - ETA: 0s - loss: 5.6017 - acc: 0.6518Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6044 - acc: 0.6516 - val_loss: 6.8726 - val_acc: 0.5078 Epoch 7/40 6475/6680 [============================>.] - ETA: 0s - loss: 5.6007 - acc: 0.6519Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 173us/step - loss: 5.6050 - acc: 0.6516 - val_loss: 6.8436 - val_acc: 0.5162 Epoch 8/40 6475/6680 [============================>.] - ETA: 0s - loss: 5.6102 - acc: 0.6514Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6046 - acc: 0.6518 - val_loss: 6.8575 - val_acc: 0.5186 Epoch 9/40 6650/6680 [============================>.] - ETA: 0s - loss: 5.5930 - acc: 0.6523Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6041 - acc: 0.6516 - val_loss: 6.8621 - val_acc: 0.5126 Epoch 10/40 6545/6680 [============================>.] - ETA: 0s - loss: 5.6014 - acc: 0.6519Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6041 - acc: 0.6518 - val_loss: 6.9093 - val_acc: 0.5126 Epoch 11/40 6545/6680 [============================>.] - ETA: 0s - loss: 5.6129 - acc: 0.6510Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 172us/step - loss: 5.6057 - acc: 0.6515 - val_loss: 6.8724 - val_acc: 0.5138 Epoch 12/40 6615/6680 [============================>.] - ETA: 0s - loss: 5.6010 - acc: 0.6522Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6020 - acc: 0.6521 - val_loss: 6.8571 - val_acc: 0.5150 Epoch 13/40 6510/6680 [============================>.] - ETA: 0s - loss: 5.5847 - acc: 0.6533Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 172us/step - loss: 5.6042 - acc: 0.6521 - val_loss: 6.8704 - val_acc: 0.5126 Epoch 14/40 6580/6680 [============================>.] - ETA: 0s - loss: 5.5992 - acc: 0.6523Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6060 - acc: 0.6518 - val_loss: 6.8789 - val_acc: 0.5054 Epoch 15/40 6650/6680 [============================>.] - ETA: 0s - loss: 5.6169 - acc: 0.6511Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6062 - acc: 0.6518 - val_loss: 6.8822 - val_acc: 0.5138 Epoch 16/40 6615/6680 [============================>.] - ETA: 0s - loss: 5.5953 - acc: 0.6523Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 177us/step - loss: 5.6060 - acc: 0.6516 - val_loss: 6.8861 - val_acc: 0.5066 Epoch 17/40 6545/6680 [============================>.] - ETA: 0s - loss: 5.6090 - acc: 0.6515Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6042 - acc: 0.6518 - val_loss: 6.9073 - val_acc: 0.5090 Epoch 18/40 6580/6680 [============================>.] - ETA: 0s - loss: 5.6050 - acc: 0.6515Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 169us/step - loss: 5.6055 - acc: 0.6515 - val_loss: 6.8782 - val_acc: 0.5102 Epoch 19/40 6510/6680 [============================>.] - ETA: 0s - loss: 5.6112 - acc: 0.6513Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6057 - acc: 0.6515 - val_loss: 6.8843 - val_acc: 0.5114 Epoch 20/40 6545/6680 [============================>.] - ETA: 0s - loss: 5.6228 - acc: 0.6506Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6033 - acc: 0.6518 - val_loss: 6.8778 - val_acc: 0.5114 Epoch 21/40 6475/6680 [============================>.] - ETA: 0s - loss: 5.5936 - acc: 0.6522Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6053 - acc: 0.6515 - val_loss: 6.8788 - val_acc: 0.5042 Epoch 22/40 6475/6680 [============================>.] - ETA: 0s - loss: 5.5881 - acc: 0.6527Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6048 - acc: 0.6516 - val_loss: 6.8894 - val_acc: 0.5102 Epoch 23/40 6650/6680 [============================>.] - ETA: 0s - loss: 5.6037 - acc: 0.6517Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6051 - acc: 0.6516 - val_loss: 6.8899 - val_acc: 0.5006 Epoch 24/40 6580/6680 [============================>.] - ETA: 0s - loss: 5.5978 - acc: 0.6524Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6057 - acc: 0.6519 - val_loss: 6.8857 - val_acc: 0.5114 Epoch 25/40 6545/6680 [============================>.] - ETA: 0s - loss: 5.5926 - acc: 0.6527Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6051 - acc: 0.6519 - val_loss: 6.8383 - val_acc: 0.5138 Epoch 26/40 6650/6680 [============================>.] - ETA: 0s - loss: 5.6116 - acc: 0.6513Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6057 - acc: 0.6516 - val_loss: 6.8608 - val_acc: 0.5114 Epoch 27/40 6650/6680 [============================>.] - ETA: 0s - loss: 5.6150 - acc: 0.6511Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6042 - acc: 0.6518 - val_loss: 6.8719 - val_acc: 0.5114 Epoch 28/40 6510/6680 [============================>.] - ETA: 0s - loss: 5.5822 - acc: 0.6531Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6057 - acc: 0.6516 - val_loss: 6.8608 - val_acc: 0.5114 Epoch 29/40 6440/6680 [===========================>..] - ETA: 0s - loss: 5.5874 - acc: 0.6530Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 1s 172us/step - loss: 5.6038 - acc: 0.6519 - val_loss: 6.8476 - val_acc: 0.5126 Epoch 30/40 6650/6680 [============================>.] - ETA: 0s - loss: 5.6186 - acc: 0.6510Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6055 - acc: 0.6518 - val_loss: 6.8617 - val_acc: 0.5186 Epoch 31/40 6335/6680 [===========================>..] - ETA: 0s - loss: 5.5899 - acc: 0.6526Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6052 - acc: 0.6516 - val_loss: 6.8579 - val_acc: 0.5114 Epoch 32/40 6615/6680 [============================>.] - ETA: 0s - loss: 5.6110 - acc: 0.6514Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6047 - acc: 0.6518 - val_loss: 6.8823 - val_acc: 0.5066 Epoch 33/40 6650/6680 [============================>.] - ETA: 0s - loss: 5.6091 - acc: 0.6514Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6056 - acc: 0.6516 - val_loss: 6.8773 - val_acc: 0.5102 Epoch 34/40 6475/6680 [============================>.] - ETA: 0s - loss: 5.6004 - acc: 0.6520Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6046 - acc: 0.6518 - val_loss: 6.8858 - val_acc: 0.5102 Epoch 35/40 6615/6680 [============================>.] - ETA: 0s - loss: 5.6186 - acc: 0.6509Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6049 - acc: 0.6518 - val_loss: 6.8673 - val_acc: 0.5066 Epoch 36/40 6615/6680 [============================>.] - ETA: 0s - loss: 5.6047 - acc: 0.6519Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6057 - acc: 0.6518 - val_loss: 6.8521 - val_acc: 0.5090 Epoch 37/40 6650/6680 [============================>.] - ETA: 0s - loss: 5.6090 - acc: 0.6516Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6031 - acc: 0.6519 - val_loss: 6.8998 - val_acc: 0.5030 Epoch 38/40 6615/6680 [============================>.] - ETA: 0s - loss: 5.6020 - acc: 0.6517Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6054 - acc: 0.6515 - val_loss: 6.8954 - val_acc: 0.5078 Epoch 39/40 6615/6680 [============================>.] - ETA: 0s - loss: 5.6110 - acc: 0.6512Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6046 - acc: 0.6516 - val_loss: 6.8740 - val_acc: 0.5114 Epoch 40/40 6545/6680 [============================>.] - ETA: 0s - loss: 5.6206 - acc: 0.6509Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6036 - acc: 0.6519 - val_loss: 6.9101 - val_acc: 0.5102 Batch size=35 Epoch=50 Train on 6680 samples, validate on 835 samples Epoch 1/50 6545/6680 [============================>.] - ETA: 0s - loss: 5.5983 - acc: 0.6519Epoch 00001: val_loss improved from inf to 6.84077, saving model to saved_models1/weights.best.vgg16_bs35_ep50.hdf5 6680/6680 [==============================] - 1s 173us/step - loss: 5.6058 - acc: 0.6515 - val_loss: 6.8408 - val_acc: 0.5162 Epoch 2/50 6545/6680 [============================>.] - ETA: 0s - loss: 5.5869 - acc: 0.6530Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6019 - acc: 0.6521 - val_loss: 6.8734 - val_acc: 0.5090 Epoch 3/50 6615/6680 [============================>.] - ETA: 0s - loss: 5.6049 - acc: 0.6519Epoch 00003: val_loss improved from 6.84077 to 6.80713, saving model to saved_models1/weights.best.vgg16_bs35_ep50.hdf5 6680/6680 [==============================] - 1s 171us/step - loss: 5.6034 - acc: 0.6519 - val_loss: 6.8071 - val_acc: 0.5198 Epoch 4/50 6580/6680 [============================>.] - ETA: 0s - loss: 5.5968 - acc: 0.6526Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6047 - acc: 0.6521 - val_loss: 6.8701 - val_acc: 0.5090 Epoch 5/50 6545/6680 [============================>.] - ETA: 0s - loss: 5.5967 - acc: 0.6523Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6043 - acc: 0.6518 - val_loss: 6.8701 - val_acc: 0.5126 Epoch 6/50 6335/6680 [===========================>..] - ETA: 0s - loss: 5.5780 - acc: 0.6537Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 169us/step - loss: 5.6036 - acc: 0.6521 - val_loss: 6.8549 - val_acc: 0.5114 Epoch 7/50 6580/6680 [============================>.] - ETA: 0s - loss: 5.6039 - acc: 0.6521Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6037 - acc: 0.6521 - val_loss: 6.8582 - val_acc: 0.5114 Epoch 8/50 6545/6680 [============================>.] - ETA: 0s - loss: 5.6049 - acc: 0.6519Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6051 - acc: 0.6519 - val_loss: 6.8512 - val_acc: 0.5114 Epoch 9/50 6650/6680 [============================>.] - ETA: 0s - loss: 5.5948 - acc: 0.6523Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6035 - acc: 0.6518 - val_loss: 6.8795 - val_acc: 0.5102 Epoch 10/50 6615/6680 [============================>.] - ETA: 0s - loss: 5.6023 - acc: 0.6517Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6057 - acc: 0.6515 - val_loss: 6.8896 - val_acc: 0.5078 Epoch 11/50 6650/6680 [============================>.] - ETA: 0s - loss: 5.5976 - acc: 0.6522Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6038 - acc: 0.6518 - val_loss: 6.8690 - val_acc: 0.5030 Epoch 12/50 6580/6680 [============================>.] - ETA: 0s - loss: 5.5942 - acc: 0.6524Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6046 - acc: 0.6518 - val_loss: 6.8813 - val_acc: 0.5090 Epoch 13/50 6440/6680 [===========================>..] - ETA: 0s - loss: 5.6264 - acc: 0.6506Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 172us/step - loss: 5.6052 - acc: 0.6519 - val_loss: 6.8874 - val_acc: 0.5114 Epoch 14/50 6510/6680 [============================>.] - ETA: 0s - loss: 5.6155 - acc: 0.6513Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 172us/step - loss: 5.6046 - acc: 0.6519 - val_loss: 6.8736 - val_acc: 0.5066 Epoch 15/50 6580/6680 [============================>.] - ETA: 0s - loss: 5.6104 - acc: 0.6515Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6060 - acc: 0.6518 - val_loss: 6.8661 - val_acc: 0.5126 Epoch 16/50 6545/6680 [============================>.] - ETA: 0s - loss: 5.6011 - acc: 0.6523Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6038 - acc: 0.6521 - val_loss: 6.9027 - val_acc: 0.5030 Epoch 17/50 6615/6680 [============================>.] - ETA: 0s - loss: 5.6168 - acc: 0.6508Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6056 - acc: 0.6515 - val_loss: 6.9067 - val_acc: 0.5018 Epoch 18/50 6615/6680 [============================>.] - ETA: 0s - loss: 5.5949 - acc: 0.6525Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6032 - acc: 0.6519 - val_loss: 6.8699 - val_acc: 0.5078 Epoch 19/50 6475/6680 [============================>.] - ETA: 0s - loss: 5.5707 - acc: 0.6541Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6024 - acc: 0.6521 - val_loss: 6.9001 - val_acc: 0.5078 Epoch 20/50 6335/6680 [===========================>..] - ETA: 0s - loss: 5.6186 - acc: 0.6511Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 169us/step - loss: 5.6034 - acc: 0.6521 - val_loss: 6.8629 - val_acc: 0.5054 Epoch 21/50 6615/6680 [============================>.] - ETA: 0s - loss: 5.6109 - acc: 0.6512Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6046 - acc: 0.6516 - val_loss: 6.8854 - val_acc: 0.5054 Epoch 22/50 6440/6680 [===========================>..] - ETA: 0s - loss: 5.5993 - acc: 0.6519Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6056 - acc: 0.6515 - val_loss: 6.8920 - val_acc: 0.5078 Epoch 23/50 6650/6680 [============================>.] - ETA: 0s - loss: 5.5926 - acc: 0.6526Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 1s 169us/step - loss: 5.6037 - acc: 0.6519 - val_loss: 6.9049 - val_acc: 0.5054 Epoch 24/50 6580/6680 [============================>.] - ETA: 0s - loss: 5.5950 - acc: 0.6524Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6030 - acc: 0.6519 - val_loss: 6.8917 - val_acc: 0.5030 Epoch 25/50 6545/6680 [============================>.] - ETA: 0s - loss: 5.6312 - acc: 0.6501Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6042 - acc: 0.6518 - val_loss: 6.8598 - val_acc: 0.5138 Epoch 26/50 6650/6680 [============================>.] - ETA: 0s - loss: 5.6091 - acc: 0.6513Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6056 - acc: 0.6515 - val_loss: 6.8613 - val_acc: 0.5102 Epoch 27/50 6580/6680 [============================>.] - ETA: 0s - loss: 5.5894 - acc: 0.6527Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6046 - acc: 0.6518 - val_loss: 6.8220 - val_acc: 0.5138 Epoch 28/50 6650/6680 [============================>.] - ETA: 0s - loss: 5.6028 - acc: 0.6519Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6042 - acc: 0.6518 - val_loss: 6.8744 - val_acc: 0.5054 Epoch 29/50 6580/6680 [============================>.] - ETA: 0s - loss: 5.6110 - acc: 0.6512Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6042 - acc: 0.6516 - val_loss: 6.8616 - val_acc: 0.5126 Epoch 30/50 6615/6680 [============================>.] - ETA: 0s - loss: 5.6024 - acc: 0.6519Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6057 - acc: 0.6516 - val_loss: 6.8862 - val_acc: 0.5090 Epoch 31/50 6650/6680 [============================>.] - ETA: 0s - loss: 5.6036 - acc: 0.6519Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6049 - acc: 0.6518 - val_loss: 6.9106 - val_acc: 0.5054 Epoch 32/50 6510/6680 [============================>.] - ETA: 0s - loss: 5.6059 - acc: 0.6516Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6032 - acc: 0.6518 - val_loss: 6.9040 - val_acc: 0.5006 Epoch 33/50 6545/6680 [============================>.] - ETA: 0s - loss: 5.5877 - acc: 0.6529Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6027 - acc: 0.6519 - val_loss: 6.8928 - val_acc: 0.5078 Epoch 34/50 6510/6680 [============================>.] - ETA: 0s - loss: 5.6187 - acc: 0.6508Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6036 - acc: 0.6518 - val_loss: 6.8671 - val_acc: 0.5138 Epoch 35/50 6580/6680 [============================>.] - ETA: 0s - loss: 5.6139 - acc: 0.6509Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6047 - acc: 0.6515 - val_loss: 6.8647 - val_acc: 0.5090 Epoch 36/50 6510/6680 [============================>.] - ETA: 0s - loss: 5.6071 - acc: 0.6515Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6044 - acc: 0.6516 - val_loss: 6.8632 - val_acc: 0.5102 Epoch 37/50 6475/6680 [============================>.] - ETA: 0s - loss: 5.5964 - acc: 0.6525Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6057 - acc: 0.6519 - val_loss: 6.8559 - val_acc: 0.5114 Epoch 38/50 6580/6680 [============================>.] - ETA: 0s - loss: 5.6329 - acc: 0.6500Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6041 - acc: 0.6518 - val_loss: 6.8995 - val_acc: 0.5030 Epoch 39/50 6615/6680 [============================>.] - ETA: 0s - loss: 5.6146 - acc: 0.6511Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6034 - acc: 0.6518 - val_loss: 6.8703 - val_acc: 0.5126 Epoch 40/50 6475/6680 [============================>.] - ETA: 0s - loss: 5.6194 - acc: 0.6510Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6052 - acc: 0.6518 - val_loss: 6.8981 - val_acc: 0.5078 Epoch 41/50 6335/6680 [===========================>..] - ETA: 0s - loss: 5.5844 - acc: 0.6530Epoch 00041: val_loss did not improve 6680/6680 [==============================] - 1s 173us/step - loss: 5.6049 - acc: 0.6518 - val_loss: 6.8701 - val_acc: 0.5066 Epoch 42/50 6650/6680 [============================>.] - ETA: 0s - loss: 5.6058 - acc: 0.6516Epoch 00042: val_loss did not improve 6680/6680 [==============================] - 1s 169us/step - loss: 5.6048 - acc: 0.6516 - val_loss: 6.8773 - val_acc: 0.5054 Epoch 43/50 6615/6680 [============================>.] - ETA: 0s - loss: 5.5791 - acc: 0.6534Epoch 00043: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6044 - acc: 0.6518 - val_loss: 6.8529 - val_acc: 0.5138 Epoch 44/50 6545/6680 [============================>.] - ETA: 0s - loss: 5.6259 - acc: 0.6504Epoch 00044: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6039 - acc: 0.6518 - val_loss: 6.8792 - val_acc: 0.5102 Epoch 45/50 6510/6680 [============================>.] - ETA: 0s - loss: 5.6362 - acc: 0.6499Epoch 00045: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6037 - acc: 0.6519 - val_loss: 6.8703 - val_acc: 0.5090 Epoch 46/50 6335/6680 [===========================>..] - ETA: 0s - loss: 5.6120 - acc: 0.6513Epoch 00046: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6035 - acc: 0.6518 - val_loss: 6.8932 - val_acc: 0.5018 Epoch 47/50 6580/6680 [============================>.] - ETA: 0s - loss: 5.6138 - acc: 0.6515Epoch 00047: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6029 - acc: 0.6521 - val_loss: 6.8562 - val_acc: 0.5126 Epoch 48/50 6615/6680 [============================>.] - ETA: 0s - loss: 5.5972 - acc: 0.6522Epoch 00048: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6031 - acc: 0.6518 - val_loss: 6.8408 - val_acc: 0.5126 Epoch 49/50 6510/6680 [============================>.] - ETA: 0s - loss: 5.6212 - acc: 0.6507Epoch 00049: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6036 - acc: 0.6518 - val_loss: 6.8382 - val_acc: 0.5150 Epoch 50/50 6545/6680 [============================>.] - ETA: 0s - loss: 5.6010 - acc: 0.6523Epoch 00050: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6021 - acc: 0.6521 - val_loss: 6.8866 - val_acc: 0.5066 Batch size=37 Epoch=20 Train on 6680 samples, validate on 835 samples Epoch 1/20 6401/6680 [===========================>..] - ETA: 0s - loss: 5.6079 - acc: 0.6515Epoch 00001: val_loss improved from inf to 6.85845, saving model to saved_models1/weights.best.vgg16_bs37_ep20.hdf5 6680/6680 [==============================] - 1s 168us/step - loss: 5.6054 - acc: 0.6516 - val_loss: 6.8584 - val_acc: 0.5078 Epoch 2/20 6623/6680 [============================>.] - ETA: 0s - loss: 5.6197 - acc: 0.6511Epoch 00002: val_loss improved from 6.85845 to 6.85738, saving model to saved_models1/weights.best.vgg16_bs37_ep20.hdf5 6680/6680 [==============================] - 1s 166us/step - loss: 5.6056 - acc: 0.6519 - val_loss: 6.8574 - val_acc: 0.5138 Epoch 3/20 6660/6680 [============================>.] - ETA: 0s - loss: 5.6088 - acc: 0.6515Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6041 - acc: 0.6518 - val_loss: 6.8861 - val_acc: 0.5126 Epoch 4/20 6586/6680 [============================>.] - ETA: 0s - loss: 5.5898 - acc: 0.6527Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 1s 166us/step - loss: 5.6037 - acc: 0.6518 - val_loss: 6.8618 - val_acc: 0.5138 Epoch 5/20 6512/6680 [============================>.] - ETA: 0s - loss: 5.6025 - acc: 0.6522Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6034 - acc: 0.6521 - val_loss: 6.8938 - val_acc: 0.5042 Epoch 6/20 6586/6680 [============================>.] - ETA: 0s - loss: 5.5996 - acc: 0.6521Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 166us/step - loss: 5.6040 - acc: 0.6518 - val_loss: 6.8888 - val_acc: 0.5102 Epoch 7/20 6549/6680 [============================>.] - ETA: 0s - loss: 5.5648 - acc: 0.6543Epoch 00007: val_loss improved from 6.85738 to 6.85246, saving model to saved_models1/weights.best.vgg16_bs37_ep20.hdf5 6680/6680 [==============================] - 1s 168us/step - loss: 5.6029 - acc: 0.6519 - val_loss: 6.8525 - val_acc: 0.5126 Epoch 8/20 6364/6680 [===========================>..] - ETA: 0s - loss: 5.6037 - acc: 0.6518Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6024 - acc: 0.6518 - val_loss: 6.9009 - val_acc: 0.5066 Epoch 9/20 6438/6680 [===========================>..] - ETA: 0s - loss: 5.6001 - acc: 0.6519Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6048 - acc: 0.6516 - val_loss: 6.8913 - val_acc: 0.5042 Epoch 10/20 6623/6680 [============================>.] - ETA: 0s - loss: 5.5961 - acc: 0.6524Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6014 - acc: 0.6521 - val_loss: 6.9158 - val_acc: 0.5078 Epoch 11/20 6660/6680 [============================>.] - ETA: 0s - loss: 5.6017 - acc: 0.6520Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6043 - acc: 0.6518 - val_loss: 6.8586 - val_acc: 0.5162 Epoch 12/20 6660/6680 [============================>.] - ETA: 0s - loss: 5.5972 - acc: 0.6523Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 166us/step - loss: 5.6022 - acc: 0.6519 - val_loss: 6.8978 - val_acc: 0.5114 Epoch 13/20 6364/6680 [===========================>..] - ETA: 0s - loss: 5.6124 - acc: 0.6515Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6059 - acc: 0.6518 - val_loss: 6.8885 - val_acc: 0.5102 Epoch 14/20 6660/6680 [============================>.] - ETA: 0s - loss: 5.6026 - acc: 0.6523Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 166us/step - loss: 5.6051 - acc: 0.6521 - val_loss: 6.8668 - val_acc: 0.5126 Epoch 15/20 6512/6680 [============================>.] - ETA: 0s - loss: 5.6268 - acc: 0.6506Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 166us/step - loss: 5.6060 - acc: 0.6519 - val_loss: 6.8647 - val_acc: 0.5066 Epoch 16/20 6401/6680 [===========================>..] - ETA: 0s - loss: 5.5849 - acc: 0.6533Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6050 - acc: 0.6521 - val_loss: 6.8759 - val_acc: 0.5114 Epoch 17/20 6327/6680 [===========================>..] - ETA: 0s - loss: 5.6038 - acc: 0.6518Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6058 - acc: 0.6516 - val_loss: 6.8585 - val_acc: 0.5150 Epoch 18/20 6549/6680 [============================>.] - ETA: 0s - loss: 5.6468 - acc: 0.6493Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 166us/step - loss: 5.6060 - acc: 0.6518 - val_loss: 6.8546 - val_acc: 0.5138 Epoch 19/20 6327/6680 [===========================>..] - ETA: 0s - loss: 5.5680 - acc: 0.6540Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6043 - acc: 0.6518 - val_loss: 6.8596 - val_acc: 0.5090 Epoch 20/20 6549/6680 [============================>.] - ETA: 0s - loss: 5.5890 - acc: 0.6529Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 166us/step - loss: 5.6025 - acc: 0.6521 - val_loss: 6.8708 - val_acc: 0.5138 Batch size=37 Epoch=35 Train on 6680 samples, validate on 835 samples Epoch 1/35 6623/6680 [============================>.] - ETA: 0s - loss: 5.5991 - acc: 0.6520Epoch 00001: val_loss improved from inf to 6.88616, saving model to saved_models1/weights.best.vgg16_bs37_ep35.hdf5 6680/6680 [==============================] - 1s 166us/step - loss: 5.6044 - acc: 0.6516 - val_loss: 6.8862 - val_acc: 0.5030 Epoch 2/35 6660/6680 [============================>.] - ETA: 0s - loss: 5.6101 - acc: 0.6517Epoch 00002: val_loss improved from 6.88616 to 6.86639, saving model to saved_models1/weights.best.vgg16_bs37_ep35.hdf5 6680/6680 [==============================] - 1s 166us/step - loss: 5.6053 - acc: 0.6519 - val_loss: 6.8664 - val_acc: 0.5018 Epoch 3/35 6623/6680 [============================>.] - ETA: 0s - loss: 5.6047 - acc: 0.6518Epoch 00003: val_loss improved from 6.86639 to 6.85749, saving model to saved_models1/weights.best.vgg16_bs37_ep35.hdf5 6680/6680 [==============================] - 1s 166us/step - loss: 5.6051 - acc: 0.6518 - val_loss: 6.8575 - val_acc: 0.5150 Epoch 4/35 6512/6680 [============================>.] - ETA: 0s - loss: 5.6114 - acc: 0.6514Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6054 - acc: 0.6518 - val_loss: 6.8742 - val_acc: 0.5114 Epoch 5/35 6401/6680 [===========================>..] - ETA: 0s - loss: 5.6308 - acc: 0.6502Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6035 - acc: 0.6518 - val_loss: 6.8755 - val_acc: 0.5078 Epoch 6/35 6327/6680 [===========================>..] - ETA: 0s - loss: 5.6056 - acc: 0.6520Epoch 00006: val_loss improved from 6.85749 to 6.85091, saving model to saved_models1/weights.best.vgg16_bs37_ep35.hdf5 6680/6680 [==============================] - 1s 166us/step - loss: 5.6038 - acc: 0.6521 - val_loss: 6.8509 - val_acc: 0.5126 Epoch 7/35 6401/6680 [===========================>..] - ETA: 0s - loss: 5.5831 - acc: 0.6530Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6033 - acc: 0.6518 - val_loss: 6.8586 - val_acc: 0.5090 Epoch 8/35 6327/6680 [===========================>..] - ETA: 0s - loss: 5.5835 - acc: 0.6532Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6045 - acc: 0.6519 - val_loss: 6.8597 - val_acc: 0.5138 Epoch 9/35 6660/6680 [============================>.] - ETA: 0s - loss: 5.6036 - acc: 0.6520Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6062 - acc: 0.6518 - val_loss: 6.8703 - val_acc: 0.5162 Epoch 10/35 6660/6680 [============================>.] - ETA: 0s - loss: 5.6007 - acc: 0.6521Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6057 - acc: 0.6518 - val_loss: 6.8859 - val_acc: 0.5174 Epoch 11/35 6401/6680 [===========================>..] - ETA: 0s - loss: 5.6065 - acc: 0.6516Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6040 - acc: 0.6518 - val_loss: 6.8935 - val_acc: 0.5090 Epoch 12/35 6401/6680 [===========================>..] - ETA: 0s - loss: 5.6238 - acc: 0.6505Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6061 - acc: 0.6516 - val_loss: 6.8840 - val_acc: 0.5090 Epoch 13/35 6475/6680 [============================>.] - ETA: 0s - loss: 5.5929 - acc: 0.6525Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 163us/step - loss: 5.6060 - acc: 0.6516 - val_loss: 6.8684 - val_acc: 0.5126 Epoch 14/35 6401/6680 [===========================>..] - ETA: 0s - loss: 5.6196 - acc: 0.6507Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6045 - acc: 0.6516 - val_loss: 6.8918 - val_acc: 0.5090 Epoch 15/35 6438/6680 [===========================>..] - ETA: 0s - loss: 5.5948 - acc: 0.6524Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 168us/step - loss: 5.6057 - acc: 0.6516 - val_loss: 6.8964 - val_acc: 0.5114 Epoch 16/35 6364/6680 [===========================>..] - ETA: 0s - loss: 5.6148 - acc: 0.6508Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6050 - acc: 0.6515 - val_loss: 6.8777 - val_acc: 0.5078 Epoch 17/35 6364/6680 [===========================>..] - ETA: 0s - loss: 5.5660 - acc: 0.6540Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6043 - acc: 0.6515 - val_loss: 6.8608 - val_acc: 0.5150 Epoch 18/35 6364/6680 [===========================>..] - ETA: 0s - loss: 5.5882 - acc: 0.6526Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6038 - acc: 0.6516 - val_loss: 6.8510 - val_acc: 0.5102 Epoch 19/35 6364/6680 [===========================>..] - ETA: 0s - loss: 5.5801 - acc: 0.6532Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6056 - acc: 0.6516 - val_loss: 6.8669 - val_acc: 0.5174 Epoch 20/35 6364/6680 [===========================>..] - ETA: 0s - loss: 5.6337 - acc: 0.6499Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6049 - acc: 0.6516 - val_loss: 6.8805 - val_acc: 0.5138 Epoch 21/35 6364/6680 [===========================>..] - ETA: 0s - loss: 5.6012 - acc: 0.6519Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6040 - acc: 0.6518 - val_loss: 6.9037 - val_acc: 0.5102 Epoch 22/35 6364/6680 [===========================>..] - ETA: 0s - loss: 5.6178 - acc: 0.6508Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6030 - acc: 0.6518 - val_loss: 6.8594 - val_acc: 0.5054 Epoch 23/35 6364/6680 [===========================>..] - ETA: 0s - loss: 5.6076 - acc: 0.6515Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6053 - acc: 0.6516 - val_loss: 6.8924 - val_acc: 0.5078 Epoch 24/35 6475/6680 [============================>.] - ETA: 0s - loss: 5.5788 - acc: 0.6536Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 1s 169us/step - loss: 5.6055 - acc: 0.6519 - val_loss: 6.8872 - val_acc: 0.5066 Epoch 25/35 6364/6680 [===========================>..] - ETA: 0s - loss: 5.5878 - acc: 0.6527Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6058 - acc: 0.6516 - val_loss: 6.8577 - val_acc: 0.5114 Epoch 26/35 6364/6680 [===========================>..] - ETA: 0s - loss: 5.6017 - acc: 0.6519Epoch 00026: val_loss improved from 6.85091 to 6.84878, saving model to saved_models1/weights.best.vgg16_bs37_ep35.hdf5 6680/6680 [==============================] - 1s 172us/step - loss: 5.6045 - acc: 0.6518 - val_loss: 6.8488 - val_acc: 0.5138 Epoch 27/35 6364/6680 [===========================>..] - ETA: 0s - loss: 5.6078 - acc: 0.6518Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6047 - acc: 0.6519 - val_loss: 6.8853 - val_acc: 0.5102 Epoch 28/35 6364/6680 [===========================>..] - ETA: 0s - loss: 5.6173 - acc: 0.6510Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6049 - acc: 0.6518 - val_loss: 6.8555 - val_acc: 0.5126 Epoch 29/35 6586/6680 [============================>.] - ETA: 0s - loss: 5.6138 - acc: 0.6512Epoch 00029: val_loss improved from 6.84878 to 6.83731, saving model to saved_models1/weights.best.vgg16_bs37_ep35.hdf5 6680/6680 [==============================] - 1s 168us/step - loss: 5.6048 - acc: 0.6518 - val_loss: 6.8373 - val_acc: 0.5126 Epoch 30/35 6549/6680 [============================>.] - ETA: 0s - loss: 5.6461 - acc: 0.6494Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6053 - acc: 0.6519 - val_loss: 6.8802 - val_acc: 0.5138 Epoch 31/35 6660/6680 [============================>.] - ETA: 0s - loss: 5.6130 - acc: 0.6511Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6034 - acc: 0.6516 - val_loss: 6.8619 - val_acc: 0.5150 Epoch 32/35 6364/6680 [===========================>..] - ETA: 0s - loss: 5.5922 - acc: 0.6527Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6026 - acc: 0.6519 - val_loss: 6.8670 - val_acc: 0.5102 Epoch 33/35 6438/6680 [===========================>..] - ETA: 0s - loss: 5.6242 - acc: 0.6507- ETA: 0s - loss: 5.9Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6038 - acc: 0.6519 - val_loss: 6.8584 - val_acc: 0.5102 Epoch 34/35 6623/6680 [============================>.] - ETA: 0s - loss: 5.6144 - acc: 0.6514Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6051 - acc: 0.6519 - val_loss: 6.8816 - val_acc: 0.5102 Epoch 35/35 6327/6680 [===========================>..] - ETA: 0s - loss: 5.6143 - acc: 0.6512Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6047 - acc: 0.6518 - val_loss: 6.8800 - val_acc: 0.5090 Batch size=37 Epoch=40 Train on 6680 samples, validate on 835 samples Epoch 1/40 6549/6680 [============================>.] - ETA: 0s - loss: 5.6262 - acc: 0.6505Epoch 00001: val_loss improved from inf to 6.86902, saving model to saved_models1/weights.best.vgg16_bs37_ep40.hdf5 6680/6680 [==============================] - 1s 166us/step - loss: 5.6045 - acc: 0.6518 - val_loss: 6.8690 - val_acc: 0.5090 Epoch 2/40 6586/6680 [============================>.] - ETA: 0s - loss: 5.5878 - acc: 0.6526Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6033 - acc: 0.6516 - val_loss: 6.8778 - val_acc: 0.5066 Epoch 3/40 6512/6680 [============================>.] - ETA: 0s - loss: 5.6036 - acc: 0.6520Epoch 00003: val_loss improved from 6.86902 to 6.85879, saving model to saved_models1/weights.best.vgg16_bs37_ep40.hdf5 6680/6680 [==============================] - 1s 165us/step - loss: 5.6050 - acc: 0.6519 - val_loss: 6.8588 - val_acc: 0.5090 Epoch 4/40 6549/6680 [============================>.] - ETA: 0s - loss: 5.6121 - acc: 0.6512Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6058 - acc: 0.6516 - val_loss: 6.8614 - val_acc: 0.5102 Epoch 5/40 6623/6680 [============================>.] - ETA: 0s - loss: 5.5936 - acc: 0.6526Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6038 - acc: 0.6519 - val_loss: 6.8831 - val_acc: 0.5150 Epoch 6/40 6364/6680 [===========================>..] - ETA: 0s - loss: 5.5918 - acc: 0.6526Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6047 - acc: 0.6518 - val_loss: 6.8850 - val_acc: 0.5102 Epoch 7/40 6549/6680 [============================>.] - ETA: 0s - loss: 5.6044 - acc: 0.6519Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 167us/step - loss: 5.6054 - acc: 0.6518 - val_loss: 6.8754 - val_acc: 0.5090 Epoch 8/40 6549/6680 [============================>.] - ETA: 0s - loss: 5.5996 - acc: 0.6520Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 166us/step - loss: 5.6056 - acc: 0.6516 - val_loss: 6.8677 - val_acc: 0.5090 Epoch 9/40 6364/6680 [===========================>..] - ETA: 0s - loss: 5.6084 - acc: 0.6516Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6037 - acc: 0.6519 - val_loss: 6.8745 - val_acc: 0.5042 Epoch 10/40 6660/6680 [============================>.] - ETA: 0s - loss: 5.6022 - acc: 0.6523Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6048 - acc: 0.6521 - val_loss: 6.8702 - val_acc: 0.5042 Epoch 11/40 6660/6680 [============================>.] - ETA: 0s - loss: 5.6027 - acc: 0.6518Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6028 - acc: 0.6518 - val_loss: 6.8911 - val_acc: 0.5114 Epoch 12/40 6401/6680 [===========================>..] - ETA: 0s - loss: 5.5923 - acc: 0.6529Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6037 - acc: 0.6519 - val_loss: 6.8705 - val_acc: 0.5066 Epoch 13/40 6327/6680 [===========================>..] - ETA: 0s - loss: 5.6217 - acc: 0.6505Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6046 - acc: 0.6516 - val_loss: 6.8815 - val_acc: 0.5042 Epoch 14/40 6438/6680 [===========================>..] - ETA: 0s - loss: 5.6044 - acc: 0.6521Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6040 - acc: 0.6521 - val_loss: 6.8617 - val_acc: 0.5066 Epoch 15/40 6512/6680 [============================>.] - ETA: 0s - loss: 5.5942 - acc: 0.6525Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6043 - acc: 0.6518 - val_loss: 6.8820 - val_acc: 0.5030 Epoch 16/40 6475/6680 [============================>.] - ETA: 0s - loss: 5.6108 - acc: 0.6514Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 163us/step - loss: 5.6051 - acc: 0.6518 - val_loss: 6.8703 - val_acc: 0.5054 Epoch 17/40 6549/6680 [============================>.] - ETA: 0s - loss: 5.5921 - acc: 0.6526Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6031 - acc: 0.6519 - val_loss: 6.8667 - val_acc: 0.5102 Epoch 18/40 6623/6680 [============================>.] - ETA: 0s - loss: 5.6135 - acc: 0.6512Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6042 - acc: 0.6518 - val_loss: 6.8967 - val_acc: 0.5066 Epoch 19/40 6623/6680 [============================>.] - ETA: 0s - loss: 5.6065 - acc: 0.6515Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6045 - acc: 0.6516 - val_loss: 6.9068 - val_acc: 0.5078 Epoch 20/40 6401/6680 [===========================>..] - ETA: 0s - loss: 5.6082 - acc: 0.6516Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 163us/step - loss: 5.6058 - acc: 0.6516 - val_loss: 6.8955 - val_acc: 0.5066 Epoch 21/40 6327/6680 [===========================>..] - ETA: 0s - loss: 5.5984 - acc: 0.6524Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6049 - acc: 0.6519 - val_loss: 6.8763 - val_acc: 0.5114 Epoch 22/40 6586/6680 [============================>.] - ETA: 0s - loss: 5.6141 - acc: 0.6515Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6051 - acc: 0.6521 - val_loss: 6.8680 - val_acc: 0.5102 Epoch 23/40 6512/6680 [============================>.] - ETA: 0s - loss: 5.6054 - acc: 0.6519Epoch 00023: val_loss improved from 6.85879 to 6.85426, saving model to saved_models1/weights.best.vgg16_bs37_ep40.hdf5 6680/6680 [==============================] - 1s 165us/step - loss: 5.6044 - acc: 0.6519 - val_loss: 6.8543 - val_acc: 0.5102 Epoch 24/40 6586/6680 [============================>.] - ETA: 0s - loss: 5.6068 - acc: 0.6517Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6043 - acc: 0.6518 - val_loss: 6.8756 - val_acc: 0.5066 Epoch 25/40 6623/6680 [============================>.] - ETA: 0s - loss: 5.5926 - acc: 0.6527Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6028 - acc: 0.6521 - val_loss: 6.8733 - val_acc: 0.5078 Epoch 26/40 6401/6680 [===========================>..] - ETA: 0s - loss: 5.6261 - acc: 0.6504Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6047 - acc: 0.6516 - val_loss: 6.8621 - val_acc: 0.5078 Epoch 27/40 6623/6680 [============================>.] - ETA: 0s - loss: 5.5989 - acc: 0.6520Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6018 - acc: 0.6518 - val_loss: 6.9102 - val_acc: 0.5042 Epoch 28/40 6623/6680 [============================>.] - ETA: 0s - loss: 5.6165 - acc: 0.6511Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6047 - acc: 0.6518 - val_loss: 6.8637 - val_acc: 0.5114 Epoch 29/40 6438/6680 [===========================>..] - ETA: 0s - loss: 5.6143 - acc: 0.6513Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 1s 163us/step - loss: 5.6040 - acc: 0.6519 - val_loss: 6.8624 - val_acc: 0.5114 Epoch 30/40 6364/6680 [===========================>..] - ETA: 0s - loss: 5.5868 - acc: 0.6530Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6048 - acc: 0.6519 - val_loss: 6.8784 - val_acc: 0.5102 Epoch 31/40 6623/6680 [============================>.] - ETA: 0s - loss: 5.6071 - acc: 0.6518Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6027 - acc: 0.6521 - val_loss: 6.8731 - val_acc: 0.5126 Epoch 32/40 6327/6680 [===========================>..] - ETA: 0s - loss: 5.6342 - acc: 0.6498Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6043 - acc: 0.6516 - val_loss: 6.8693 - val_acc: 0.5150 Epoch 33/40 6401/6680 [===========================>..] - ETA: 0s - loss: 5.6176 - acc: 0.6510Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6050 - acc: 0.6518 - val_loss: 6.8738 - val_acc: 0.5162 Epoch 34/40 6586/6680 [============================>.] - ETA: 0s - loss: 5.6064 - acc: 0.6517Epoch 00034: val_loss improved from 6.85426 to 6.84940, saving model to saved_models1/weights.best.vgg16_bs37_ep40.hdf5 6680/6680 [==============================] - 1s 167us/step - loss: 5.6047 - acc: 0.6518 - val_loss: 6.8494 - val_acc: 0.5150 Epoch 35/40 6364/6680 [===========================>..] - ETA: 0s - loss: 5.6316 - acc: 0.6501Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6055 - acc: 0.6516 - val_loss: 6.8601 - val_acc: 0.5138 Epoch 36/40 6401/6680 [===========================>..] - ETA: 0s - loss: 5.6036 - acc: 0.6521Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6036 - acc: 0.6521 - val_loss: 6.8944 - val_acc: 0.5102 Epoch 37/40 6475/6680 [============================>.] - ETA: 0s - loss: 5.6095 - acc: 0.6517Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6054 - acc: 0.6519 - val_loss: 6.8534 - val_acc: 0.5102 Epoch 38/40 6364/6680 [===========================>..] - ETA: 0s - loss: 5.6184 - acc: 0.6508Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6036 - acc: 0.6518 - val_loss: 6.8541 - val_acc: 0.5126 Epoch 39/40 6660/6680 [============================>.] - ETA: 0s - loss: 5.5942 - acc: 0.6523Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6040 - acc: 0.6516 - val_loss: 6.8654 - val_acc: 0.5162 Epoch 40/40 6660/6680 [============================>.] - ETA: 0s - loss: 5.6022 - acc: 0.6520Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6048 - acc: 0.6518 - val_loss: 6.8803 - val_acc: 0.5126 Batch size=37 Epoch=50 Train on 6680 samples, validate on 835 samples Epoch 1/50 6660/6680 [============================>.] - ETA: 0s - loss: 5.6017 - acc: 0.6518Epoch 00001: val_loss improved from inf to 6.87337, saving model to saved_models1/weights.best.vgg16_bs37_ep50.hdf5 6680/6680 [==============================] - 1s 166us/step - loss: 5.6042 - acc: 0.6516 - val_loss: 6.8734 - val_acc: 0.5150 Epoch 2/50 6401/6680 [===========================>..] - ETA: 0s - loss: 5.6047 - acc: 0.6518Epoch 00002: val_loss improved from 6.87337 to 6.85252, saving model to saved_models1/weights.best.vgg16_bs37_ep50.hdf5 6680/6680 [==============================] - 1s 165us/step - loss: 5.6047 - acc: 0.6518 - val_loss: 6.8525 - val_acc: 0.5162 Epoch 3/50 6660/6680 [============================>.] - ETA: 0s - loss: 5.6059 - acc: 0.6517Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6036 - acc: 0.6518 - val_loss: 6.8638 - val_acc: 0.5174 Epoch 4/50 6549/6680 [============================>.] - ETA: 0s - loss: 5.6168 - acc: 0.6509Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 1s 166us/step - loss: 5.6056 - acc: 0.6516 - val_loss: 6.8542 - val_acc: 0.5162 Epoch 5/50 6475/6680 [============================>.] - ETA: 0s - loss: 5.5918 - acc: 0.6528Epoch 00005: val_loss improved from 6.85252 to 6.85111, saving model to saved_models1/weights.best.vgg16_bs37_ep50.hdf5 6680/6680 [==============================] - 1s 165us/step - loss: 5.6060 - acc: 0.6519 - val_loss: 6.8511 - val_acc: 0.5150 Epoch 6/50 6660/6680 [============================>.] - ETA: 0s - loss: 5.6086 - acc: 0.6517Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6039 - acc: 0.6519 - val_loss: 6.8871 - val_acc: 0.5054 Epoch 7/50 6549/6680 [============================>.] - ETA: 0s - loss: 5.6047 - acc: 0.6519Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6034 - acc: 0.6519 - val_loss: 6.8702 - val_acc: 0.5102 Epoch 8/50 6660/6680 [============================>.] - ETA: 0s - loss: 5.6073 - acc: 0.6517Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6050 - acc: 0.6518 - val_loss: 6.8628 - val_acc: 0.5102 Epoch 9/50 6549/6680 [============================>.] - ETA: 0s - loss: 5.6267 - acc: 0.6506Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 166us/step - loss: 5.6057 - acc: 0.6519 - val_loss: 6.8890 - val_acc: 0.5066 Epoch 10/50 6623/6680 [============================>.] - ETA: 0s - loss: 5.6135 - acc: 0.6512Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6042 - acc: 0.6518 - val_loss: 6.8743 - val_acc: 0.5102 Epoch 11/50 6364/6680 [===========================>..] - ETA: 0s - loss: 5.6305 - acc: 0.6501Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6030 - acc: 0.6518 - val_loss: 6.8518 - val_acc: 0.5126 Epoch 12/50 6327/6680 [===========================>..] - ETA: 0s - loss: 5.6297 - acc: 0.6504Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6056 - acc: 0.6518 - val_loss: 6.8597 - val_acc: 0.5138 Epoch 13/50 6512/6680 [============================>.] - ETA: 0s - loss: 5.6051 - acc: 0.6517Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 167us/step - loss: 5.6041 - acc: 0.6518 - val_loss: 6.8617 - val_acc: 0.5066 Epoch 14/50 6586/6680 [============================>.] - ETA: 0s - loss: 5.6092 - acc: 0.6518Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6051 - acc: 0.6521 - val_loss: 6.8586 - val_acc: 0.5102 Epoch 15/50 6401/6680 [===========================>..] - ETA: 0s - loss: 5.5727 - acc: 0.6541Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6046 - acc: 0.6519 - val_loss: 6.8589 - val_acc: 0.5126 Epoch 16/50 6586/6680 [============================>.] - ETA: 0s - loss: 5.5850 - acc: 0.6531Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6053 - acc: 0.6518 - val_loss: 6.8596 - val_acc: 0.5090 Epoch 17/50 6549/6680 [============================>.] - ETA: 0s - loss: 5.5985 - acc: 0.6523Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6045 - acc: 0.6519 - val_loss: 6.8713 - val_acc: 0.5090 Epoch 18/50 6512/6680 [============================>.] - ETA: 0s - loss: 5.5891 - acc: 0.6525Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 162us/step - loss: 5.6053 - acc: 0.6515 - val_loss: 6.8699 - val_acc: 0.5138 Epoch 19/50 6401/6680 [===========================>..] - ETA: 0s - loss: 5.6144 - acc: 0.6510Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6043 - acc: 0.6516 - val_loss: 6.8532 - val_acc: 0.5126 Epoch 20/50 6512/6680 [============================>.] - ETA: 0s - loss: 5.6025 - acc: 0.6520Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 166us/step - loss: 5.6040 - acc: 0.6519 - val_loss: 6.8656 - val_acc: 0.5090 Epoch 21/50 6401/6680 [===========================>..] - ETA: 0s - loss: 5.5824 - acc: 0.6532Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6050 - acc: 0.6518 - val_loss: 6.8693 - val_acc: 0.5102 Epoch 22/50 6660/6680 [============================>.] - ETA: 0s - loss: 5.5924 - acc: 0.6527Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 1s 166us/step - loss: 5.6046 - acc: 0.6519 - val_loss: 6.8762 - val_acc: 0.5090 Epoch 23/50 6623/6680 [============================>.] - ETA: 0s - loss: 5.5996 - acc: 0.6518Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6049 - acc: 0.6515 - val_loss: 6.8752 - val_acc: 0.5114 Epoch 24/50 6512/6680 [============================>.] - ETA: 0s - loss: 5.5862 - acc: 0.6528Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6040 - acc: 0.6516 - val_loss: 6.8665 - val_acc: 0.5114 Epoch 25/50 6327/6680 [===========================>..] - ETA: 0s - loss: 5.6180 - acc: 0.6510Epoch 00025: val_loss improved from 6.85111 to 6.84701, saving model to saved_models1/weights.best.vgg16_bs37_ep50.hdf5 6680/6680 [==============================] - 1s 166us/step - loss: 5.6047 - acc: 0.6518 - val_loss: 6.8470 - val_acc: 0.5102 Epoch 26/50 6401/6680 [===========================>..] - ETA: 0s - loss: 5.6026 - acc: 0.6516Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6050 - acc: 0.6515 - val_loss: 6.8515 - val_acc: 0.5090 Epoch 27/50 6586/6680 [============================>.] - ETA: 0s - loss: 5.6025 - acc: 0.6520Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 1s 166us/step - loss: 5.6033 - acc: 0.6519 - val_loss: 6.8857 - val_acc: 0.5054 Epoch 28/50 6549/6680 [============================>.] - ETA: 0s - loss: 5.5879 - acc: 0.6528Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6038 - acc: 0.6518 - val_loss: 6.8648 - val_acc: 0.5078 Epoch 29/50 6623/6680 [============================>.] - ETA: 0s - loss: 5.6017 - acc: 0.6521Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6045 - acc: 0.6519 - val_loss: 6.8795 - val_acc: 0.5126 Epoch 30/50 6549/6680 [============================>.] - ETA: 0s - loss: 5.5873 - acc: 0.6529Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 1s 163us/step - loss: 5.6032 - acc: 0.6519 - val_loss: 6.8599 - val_acc: 0.5126 Epoch 31/50 6586/6680 [============================>.] - ETA: 0s - loss: 5.5879 - acc: 0.6529Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6057 - acc: 0.6518 - val_loss: 6.8663 - val_acc: 0.5102 Epoch 32/50 6623/6680 [============================>.] - ETA: 0s - loss: 5.6117 - acc: 0.6514Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6048 - acc: 0.6518 - val_loss: 6.8608 - val_acc: 0.5114 Epoch 33/50 6401/6680 [===========================>..] - ETA: 0s - loss: 5.6278 - acc: 0.6505Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6051 - acc: 0.6519 - val_loss: 6.8632 - val_acc: 0.5078 Epoch 34/50 6364/6680 [===========================>..] - ETA: 0s - loss: 5.5994 - acc: 0.6521Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 1s 163us/step - loss: 5.6048 - acc: 0.6518 - val_loss: 6.8730 - val_acc: 0.5090 Epoch 35/50 6364/6680 [===========================>..] - ETA: 0s - loss: 5.5908 - acc: 0.6524Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6053 - acc: 0.6515 - val_loss: 6.8830 - val_acc: 0.5078 Epoch 36/50 6401/6680 [===========================>..] - ETA: 0s - loss: 5.6273 - acc: 0.6504Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6046 - acc: 0.6518 - val_loss: 6.8658 - val_acc: 0.5078 Epoch 37/50 6438/6680 [===========================>..] - ETA: 0s - loss: 5.5717 - acc: 0.6538Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 1s 163us/step - loss: 5.6050 - acc: 0.6516 - val_loss: 6.8692 - val_acc: 0.5090 Epoch 38/50 6660/6680 [============================>.] - ETA: 0s - loss: 5.5996 - acc: 0.6521Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6045 - acc: 0.6518 - val_loss: 6.8767 - val_acc: 0.5042 Epoch 39/50 6586/6680 [============================>.] - ETA: 0s - loss: 5.5892 - acc: 0.6526Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6046 - acc: 0.6516 - val_loss: 6.8628 - val_acc: 0.5066 Epoch 40/50 6475/6680 [============================>.] - ETA: 0s - loss: 5.6199 - acc: 0.6510Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6043 - acc: 0.6519 - val_loss: 6.8755 - val_acc: 0.5042 Epoch 41/50 6586/6680 [============================>.] - ETA: 0s - loss: 5.5999 - acc: 0.6521Epoch 00041: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6056 - acc: 0.6518 - val_loss: 6.8799 - val_acc: 0.5018 Epoch 42/50 6623/6680 [============================>.] - ETA: 0s - loss: 5.6048 - acc: 0.6521Epoch 00042: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6052 - acc: 0.6521 - val_loss: 6.8781 - val_acc: 0.5006 Epoch 43/50 6438/6680 [===========================>..] - ETA: 0s - loss: 5.6117 - acc: 0.6514Epoch 00043: val_loss did not improve 6680/6680 [==============================] - 1s 163us/step - loss: 5.6039 - acc: 0.6519 - val_loss: 6.8826 - val_acc: 0.5006 Epoch 44/50 6438/6680 [===========================>..] - ETA: 0s - loss: 5.6244 - acc: 0.6504Epoch 00044: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6040 - acc: 0.6516 - val_loss: 6.8885 - val_acc: 0.5054 Epoch 45/50 6401/6680 [===========================>..] - ETA: 0s - loss: 5.6183 - acc: 0.6510Epoch 00045: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6052 - acc: 0.6518 - val_loss: 6.8818 - val_acc: 0.5066 Epoch 46/50 6438/6680 [===========================>..] - ETA: 0s - loss: 5.5737 - acc: 0.6535Epoch 00046: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6053 - acc: 0.6515 - val_loss: 6.8721 - val_acc: 0.5078 Epoch 47/50 6401/6680 [===========================>..] - ETA: 0s - loss: 5.5876 - acc: 0.6529Epoch 00047: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6052 - acc: 0.6518 - val_loss: 6.8823 - val_acc: 0.5078 Epoch 48/50 6586/6680 [============================>.] - ETA: 0s - loss: 5.5992 - acc: 0.6523Epoch 00048: val_loss did not improve 6680/6680 [==============================] - 1s 166us/step - loss: 5.6048 - acc: 0.6519 - val_loss: 6.8934 - val_acc: 0.5054 Epoch 49/50 6438/6680 [===========================>..] - ETA: 0s - loss: 5.6154 - acc: 0.6513Epoch 00049: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6026 - acc: 0.6521 - val_loss: 6.8890 - val_acc: 0.5066 Epoch 50/50 6364/6680 [===========================>..] - ETA: 0s - loss: 5.5937 - acc: 0.6526Epoch 00050: val_loss did not improve 6680/6680 [==============================] - 1s 165us/step - loss: 5.6042 - acc: 0.6519 - val_loss: 6.8951 - val_acc: 0.5066 Batch size=40 Epoch=20 Train on 6680 samples, validate on 835 samples Epoch 1/20 6520/6680 [============================>.] - ETA: 0s - loss: 5.6009 - acc: 0.6520Epoch 00001: val_loss improved from inf to 6.87440, saving model to saved_models1/weights.best.vgg16_bs40_ep20.hdf5 6680/6680 [==============================] - 1s 159us/step - loss: 5.6019 - acc: 0.6519 - val_loss: 6.8744 - val_acc: 0.5054 Epoch 2/20 6520/6680 [============================>.] - ETA: 0s - loss: 5.6258 - acc: 0.6508Epoch 00002: val_loss improved from 6.87440 to 6.86759, saving model to saved_models1/weights.best.vgg16_bs40_ep20.hdf5 6680/6680 [==============================] - 1s 158us/step - loss: 5.6037 - acc: 0.6519 - val_loss: 6.8676 - val_acc: 0.5090 Epoch 3/20 6520/6680 [============================>.] - ETA: 0s - loss: 5.6279 - acc: 0.6505Epoch 00003: val_loss improved from 6.86759 to 6.84342, saving model to saved_models1/weights.best.vgg16_bs40_ep20.hdf5 6680/6680 [==============================] - 1s 158us/step - loss: 5.6041 - acc: 0.6519 - val_loss: 6.8434 - val_acc: 0.5102 Epoch 4/20 6520/6680 [============================>.] - ETA: 0s - loss: 5.6166 - acc: 0.6511Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6051 - acc: 0.6518 - val_loss: 6.8579 - val_acc: 0.5078 Epoch 5/20 6520/6680 [============================>.] - ETA: 0s - loss: 5.5921 - acc: 0.6525Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6054 - acc: 0.6516 - val_loss: 6.8680 - val_acc: 0.5018 Epoch 6/20 6520/6680 [============================>.] - ETA: 0s - loss: 5.6334 - acc: 0.6498Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6046 - acc: 0.6516 - val_loss: 6.8579 - val_acc: 0.5042 Epoch 7/20 6520/6680 [============================>.] - ETA: 0s - loss: 5.6121 - acc: 0.6515Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6031 - acc: 0.6521 - val_loss: 6.8837 - val_acc: 0.5078 Epoch 8/20 6520/6680 [============================>.] - ETA: 0s - loss: 5.5934 - acc: 0.6526Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6042 - acc: 0.6519 - val_loss: 6.8533 - val_acc: 0.5126 Epoch 9/20 6520/6680 [============================>.] - ETA: 0s - loss: 5.6134 - acc: 0.6512Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6044 - acc: 0.6518 - val_loss: 6.8688 - val_acc: 0.5090 Epoch 10/20 6520/6680 [============================>.] - ETA: 0s - loss: 5.6048 - acc: 0.6520Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6033 - acc: 0.6521 - val_loss: 6.8796 - val_acc: 0.5054 Epoch 11/20 6520/6680 [============================>.] - ETA: 0s - loss: 5.6072 - acc: 0.6517Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6032 - acc: 0.6519 - val_loss: 6.8863 - val_acc: 0.5114 Epoch 12/20 6640/6680 [============================>.] - ETA: 0s - loss: 5.5968 - acc: 0.6523Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 164us/step - loss: 5.6043 - acc: 0.6518 - val_loss: 6.8661 - val_acc: 0.5102 Epoch 13/20 6520/6680 [============================>.] - ETA: 0s - loss: 5.6111 - acc: 0.6515Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6045 - acc: 0.6519 - val_loss: 6.8827 - val_acc: 0.5114 Epoch 14/20 6520/6680 [============================>.] - ETA: 0s - loss: 5.6076 - acc: 0.6515Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6036 - acc: 0.6518 - val_loss: 6.8951 - val_acc: 0.5078 Epoch 15/20 6480/6680 [============================>.] - ETA: 0s - loss: 5.5797 - acc: 0.6532Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6057 - acc: 0.6516 - val_loss: 6.8663 - val_acc: 0.5138 Epoch 16/20 6520/6680 [============================>.] - ETA: 0s - loss: 5.5924 - acc: 0.6526Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6037 - acc: 0.6518 - val_loss: 6.8629 - val_acc: 0.5150 Epoch 17/20 6520/6680 [============================>.] - ETA: 0s - loss: 5.5979 - acc: 0.6523Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6038 - acc: 0.6519 - val_loss: 6.8597 - val_acc: 0.5126 Epoch 18/20 6560/6680 [============================>.] - ETA: 0s - loss: 5.6093 - acc: 0.6515Epoch 00018: val_loss improved from 6.84342 to 6.84174, saving model to saved_models1/weights.best.vgg16_bs40_ep20.hdf5 6680/6680 [==============================] - 1s 157us/step - loss: 5.6051 - acc: 0.6518 - val_loss: 6.8417 - val_acc: 0.5138 Epoch 19/20 6520/6680 [============================>.] - ETA: 0s - loss: 5.5705 - acc: 0.6538Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6035 - acc: 0.6518 - val_loss: 6.8682 - val_acc: 0.5150 Epoch 20/20 6520/6680 [============================>.] - ETA: 0s - loss: 5.5919 - acc: 0.6523Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6051 - acc: 0.6515 - val_loss: 6.8459 - val_acc: 0.5138 Batch size=40 Epoch=35 Train on 6680 samples, validate on 835 samples Epoch 1/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.6213 - acc: 0.6509Epoch 00001: val_loss improved from inf to 6.85722, saving model to saved_models1/weights.best.vgg16_bs40_ep35.hdf5 6680/6680 [==============================] - 1s 159us/step - loss: 5.6049 - acc: 0.6519 - val_loss: 6.8572 - val_acc: 0.5138 Epoch 2/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.5778 - acc: 0.6537Epoch 00002: val_loss improved from 6.85722 to 6.85657, saving model to saved_models1/weights.best.vgg16_bs40_ep35.hdf5 6680/6680 [==============================] - 1s 158us/step - loss: 5.6047 - acc: 0.6519 - val_loss: 6.8566 - val_acc: 0.5138 Epoch 3/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.6149 - acc: 0.6514Epoch 00003: val_loss improved from 6.85657 to 6.84041, saving model to saved_models1/weights.best.vgg16_bs40_ep35.hdf5 6680/6680 [==============================] - 1s 158us/step - loss: 5.6059 - acc: 0.6519 - val_loss: 6.8404 - val_acc: 0.5114 Epoch 4/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.6286 - acc: 0.6503Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6048 - acc: 0.6518 - val_loss: 6.8492 - val_acc: 0.5126 Epoch 5/35 6560/6680 [============================>.] - ETA: 0s - loss: 5.6016 - acc: 0.6521Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6047 - acc: 0.6519 - val_loss: 6.8738 - val_acc: 0.5126 Epoch 6/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.6074 - acc: 0.6515Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6034 - acc: 0.6518 - val_loss: 6.8721 - val_acc: 0.5138 Epoch 7/35 6480/6680 [============================>.] - ETA: 0s - loss: 5.5914 - acc: 0.6523Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6049 - acc: 0.6515 - val_loss: 6.8715 - val_acc: 0.5114 Epoch 8/35 6560/6680 [============================>.] - ETA: 0s - loss: 5.5812 - acc: 0.6535Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6040 - acc: 0.6521 - val_loss: 6.8734 - val_acc: 0.5126 Epoch 9/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.5799 - acc: 0.6532Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6050 - acc: 0.6516 - val_loss: 6.8857 - val_acc: 0.5126 Epoch 10/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.5946 - acc: 0.6523Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6054 - acc: 0.6516 - val_loss: 6.8780 - val_acc: 0.5138 Epoch 11/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.5995 - acc: 0.6520Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6053 - acc: 0.6516 - val_loss: 6.8733 - val_acc: 0.5126 Epoch 12/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.6058 - acc: 0.6517Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6042 - acc: 0.6518 - val_loss: 6.8719 - val_acc: 0.5186 Epoch 13/35 6560/6680 [============================>.] - ETA: 0s - loss: 5.6221 - acc: 0.6508Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6055 - acc: 0.6518 - val_loss: 6.8712 - val_acc: 0.5102 Epoch 14/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.6164 - acc: 0.6509Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6050 - acc: 0.6516 - val_loss: 6.8817 - val_acc: 0.5126 Epoch 15/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.5801 - acc: 0.6531Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6053 - acc: 0.6515 - val_loss: 6.8812 - val_acc: 0.5126 Epoch 16/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.5993 - acc: 0.6520Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6051 - acc: 0.6516 - val_loss: 6.8837 - val_acc: 0.5102 Epoch 17/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.6175 - acc: 0.6514Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6036 - acc: 0.6522 - val_loss: 6.8612 - val_acc: 0.5126 Epoch 18/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.5607 - acc: 0.6548Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6058 - acc: 0.6519 - val_loss: 6.8649 - val_acc: 0.5126 Epoch 19/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.6298 - acc: 0.6505Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6060 - acc: 0.6519 - val_loss: 6.8638 - val_acc: 0.5126 Epoch 20/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.6041 - acc: 0.6518Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6050 - acc: 0.6518 - val_loss: 6.8502 - val_acc: 0.5162 Epoch 21/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.6005 - acc: 0.6523Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6042 - acc: 0.6519 - val_loss: 6.8828 - val_acc: 0.5090 Epoch 22/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.5861 - acc: 0.6532Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6043 - acc: 0.6521 - val_loss: 6.8599 - val_acc: 0.5126 Epoch 23/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.6053 - acc: 0.6515Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6037 - acc: 0.6516 - val_loss: 6.8728 - val_acc: 0.5102 Epoch 24/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.6063 - acc: 0.6518Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6023 - acc: 0.6521 - val_loss: 6.8734 - val_acc: 0.5162 Epoch 25/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.6032 - acc: 0.6517Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6042 - acc: 0.6516 - val_loss: 6.8669 - val_acc: 0.5102 Epoch 26/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.6088 - acc: 0.6517Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6047 - acc: 0.6519 - val_loss: 6.8625 - val_acc: 0.5138 Epoch 27/35 6480/6680 [============================>.] - ETA: 0s - loss: 5.5729 - acc: 0.6537Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 1s 160us/step - loss: 5.6039 - acc: 0.6518 - val_loss: 6.8727 - val_acc: 0.5114 Epoch 28/35 6560/6680 [============================>.] - ETA: 0s - loss: 5.6095 - acc: 0.6515Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6053 - acc: 0.6518 - val_loss: 6.8748 - val_acc: 0.5126 Epoch 29/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.5968 - acc: 0.6526Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6052 - acc: 0.6521 - val_loss: 6.8568 - val_acc: 0.5126 Epoch 30/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.6044 - acc: 0.6517Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6053 - acc: 0.6516 - val_loss: 6.8715 - val_acc: 0.5102 Epoch 31/35 6560/6680 [============================>.] - ETA: 0s - loss: 5.5791 - acc: 0.6532Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6043 - acc: 0.6516 - val_loss: 6.8915 - val_acc: 0.5090 Epoch 32/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.6161 - acc: 0.6512Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6046 - acc: 0.6519 - val_loss: 6.8810 - val_acc: 0.5114 Epoch 33/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.6109 - acc: 0.6515Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6030 - acc: 0.6519 - val_loss: 6.8774 - val_acc: 0.5138 Epoch 34/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.6057 - acc: 0.6518Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6041 - acc: 0.6519 - val_loss: 6.8897 - val_acc: 0.5138 Epoch 35/35 6520/6680 [============================>.] - ETA: 0s - loss: 5.6175 - acc: 0.6511Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6053 - acc: 0.6518 - val_loss: 6.8729 - val_acc: 0.5138 Batch size=40 Epoch=40 Train on 6680 samples, validate on 835 samples Epoch 1/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.5855 - acc: 0.6529Epoch 00001: val_loss improved from inf to 6.87384, saving model to saved_models1/weights.best.vgg16_bs40_ep40.hdf5 6680/6680 [==============================] - 1s 158us/step - loss: 5.6038 - acc: 0.6518 - val_loss: 6.8738 - val_acc: 0.5114 Epoch 2/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.6225 - acc: 0.6508Epoch 00002: val_loss improved from 6.87384 to 6.83667, saving model to saved_models1/weights.best.vgg16_bs40_ep40.hdf5 6680/6680 [==============================] - 1s 159us/step - loss: 5.6045 - acc: 0.6518 - val_loss: 6.8367 - val_acc: 0.5150 Epoch 3/40 6480/6680 [============================>.] - ETA: 0s - loss: 5.6034 - acc: 0.6519Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 1s 160us/step - loss: 5.6045 - acc: 0.6518 - val_loss: 6.8666 - val_acc: 0.5102 Epoch 4/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.6105 - acc: 0.6515Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6040 - acc: 0.6519 - val_loss: 6.8713 - val_acc: 0.5114 Epoch 5/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.5983 - acc: 0.6523Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6041 - acc: 0.6519 - val_loss: 6.8623 - val_acc: 0.5126 Epoch 6/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.5807 - acc: 0.6534Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6039 - acc: 0.6519 - val_loss: 6.8483 - val_acc: 0.5114 Epoch 7/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.6105 - acc: 0.6514Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6040 - acc: 0.6518 - val_loss: 6.8488 - val_acc: 0.5090 Epoch 8/40 6560/6680 [============================>.] - ETA: 0s - loss: 5.5965 - acc: 0.6523Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6046 - acc: 0.6518 - val_loss: 6.8663 - val_acc: 0.5078 Epoch 9/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.6160 - acc: 0.6511Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6045 - acc: 0.6518 - val_loss: 6.8530 - val_acc: 0.5066 Epoch 10/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.5888 - acc: 0.6526Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6046 - acc: 0.6516 - val_loss: 6.8681 - val_acc: 0.5102 Epoch 11/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.6258 - acc: 0.6503Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6045 - acc: 0.6516 - val_loss: 6.8781 - val_acc: 0.5078 Epoch 12/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.6061 - acc: 0.6518Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6046 - acc: 0.6519 - val_loss: 6.8908 - val_acc: 0.5066 Epoch 13/40 6600/6680 [============================>.] - ETA: 0s - loss: 5.6043 - acc: 0.6518Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6048 - acc: 0.6518 - val_loss: 6.8522 - val_acc: 0.5126 Epoch 14/40 6560/6680 [============================>.] - ETA: 0s - loss: 5.6048 - acc: 0.6518Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 155us/step - loss: 5.6055 - acc: 0.6518 - val_loss: 6.8582 - val_acc: 0.5102 Epoch 15/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.6143 - acc: 0.6511Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6053 - acc: 0.6516 - val_loss: 6.8598 - val_acc: 0.5102 Epoch 16/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.6087 - acc: 0.6515Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6047 - acc: 0.6518 - val_loss: 6.8732 - val_acc: 0.5114 Epoch 17/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.5731 - acc: 0.6537Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6037 - acc: 0.6518 - val_loss: 6.9067 - val_acc: 0.5090 Epoch 18/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.6002 - acc: 0.6523Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6036 - acc: 0.6521 - val_loss: 6.8628 - val_acc: 0.5102 Epoch 19/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.5948 - acc: 0.6525Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6047 - acc: 0.6518 - val_loss: 6.8764 - val_acc: 0.5078 Epoch 20/40 6560/6680 [============================>.] - ETA: 0s - loss: 5.5847 - acc: 0.6529Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6050 - acc: 0.6516 - val_loss: 6.8660 - val_acc: 0.5090 Epoch 21/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.6307 - acc: 0.6502Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6045 - acc: 0.6518 - val_loss: 6.8850 - val_acc: 0.5030 Epoch 22/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.6135 - acc: 0.6512Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6045 - acc: 0.6518 - val_loss: 6.8731 - val_acc: 0.5102 Epoch 23/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.6097 - acc: 0.6518Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6048 - acc: 0.6521 - val_loss: 6.8673 - val_acc: 0.5102 Epoch 24/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.6076 - acc: 0.6515Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6043 - acc: 0.6516 - val_loss: 6.8731 - val_acc: 0.5126 Epoch 25/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.5988 - acc: 0.6523Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6047 - acc: 0.6519 - val_loss: 6.8453 - val_acc: 0.5138 Epoch 26/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.6100 - acc: 0.6515Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6035 - acc: 0.6519 - val_loss: 6.8610 - val_acc: 0.5126 Epoch 27/40 6520/6680 [============================>.] - ETA: 0s - loss: 5.6255 - acc: 0.6505Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6053 - acc: 0.6516 - val_loss: 6.8644 - val_acc: 0.5090 Epoch 28/40 6440/6680 [===========================>..] - ETA: 0s - loss: 5.6194 - acc: 0.6508Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 1s 168us/step - loss: 5.6033 - acc: 0.6518 - val_loss: 6.8612 - val_acc: 0.5114 Epoch 29/40 6600/6680 [============================>.] - ETA: 0s - loss: 5.6087 - acc: 0.6515Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 1s 169us/step - loss: 5.6042 - acc: 0.6518 - val_loss: 6.8646 - val_acc: 0.5126 Epoch 30/40 6360/6680 [===========================>..] - ETA: 0s - loss: 5.5935 - acc: 0.6525Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 1s 167us/step - loss: 5.6055 - acc: 0.6518 - val_loss: 6.8641 - val_acc: 0.5102 Epoch 31/40 6360/6680 [===========================>..] - ETA: 0s - loss: 5.6196 - acc: 0.6509Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 1s 167us/step - loss: 5.6037 - acc: 0.6519 - val_loss: 6.9054 - val_acc: 0.5066 Epoch 32/40 6360/6680 [===========================>..] - ETA: 0s - loss: 5.6364 - acc: 0.6497Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 1s 173us/step - loss: 5.6052 - acc: 0.6516 - val_loss: 6.8873 - val_acc: 0.5066 Epoch 33/40 6400/6680 [===========================>..] - ETA: 0s - loss: 5.6109 - acc: 0.6516Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 1s 171us/step - loss: 5.6040 - acc: 0.6519 - val_loss: 6.8749 - val_acc: 0.5090 Epoch 34/40 6400/6680 [===========================>..] - ETA: 0s - loss: 5.5707 - acc: 0.6539Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 1s 167us/step - loss: 5.6026 - acc: 0.6519 - val_loss: 6.9069 - val_acc: 0.5066 Epoch 35/40 6600/6680 [============================>.] - ETA: 0s - loss: 5.5870 - acc: 0.6529Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6034 - acc: 0.6518 - val_loss: 6.8660 - val_acc: 0.5114 Epoch 36/40 6640/6680 [============================>.] - ETA: 0s - loss: 5.6016 - acc: 0.6521Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 1s 169us/step - loss: 5.6043 - acc: 0.6519 - val_loss: 6.8671 - val_acc: 0.5102 Epoch 37/40 6360/6680 [===========================>..] - ETA: 0s - loss: 5.5865 - acc: 0.6530Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 1s 167us/step - loss: 5.6036 - acc: 0.6519 - val_loss: 6.8622 - val_acc: 0.5078 Epoch 38/40 6640/6680 [============================>.] - ETA: 0s - loss: 5.5979 - acc: 0.6523Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 1s 168us/step - loss: 5.6054 - acc: 0.6518 - val_loss: 6.8664 - val_acc: 0.5138 Epoch 39/40 6400/6680 [===========================>..] - ETA: 0s - loss: 5.6100 - acc: 0.6516Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 1s 170us/step - loss: 5.6041 - acc: 0.6519 - val_loss: 6.8497 - val_acc: 0.5114 Epoch 40/40 6440/6680 [===========================>..] - ETA: 0s - loss: 5.6080 - acc: 0.6514Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 1s 167us/step - loss: 5.6044 - acc: 0.6516 - val_loss: 6.8734 - val_acc: 0.5126 Batch size=40 Epoch=50 Train on 6680 samples, validate on 835 samples Epoch 1/50 6360/6680 [===========================>..] - ETA: 0s - loss: 5.6242 - acc: 0.6508Epoch 00001: val_loss improved from inf to 6.87814, saving model to saved_models1/weights.best.vgg16_bs40_ep50.hdf5 6680/6680 [==============================] - 1s 169us/step - loss: 5.6045 - acc: 0.6519 - val_loss: 6.8781 - val_acc: 0.5114 Epoch 2/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.5702 - acc: 0.6538Epoch 00002: val_loss improved from 6.87814 to 6.85458, saving model to saved_models1/weights.best.vgg16_bs40_ep50.hdf5 6680/6680 [==============================] - 1s 159us/step - loss: 5.6033 - acc: 0.6518 - val_loss: 6.8546 - val_acc: 0.5150 Epoch 3/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6078 - acc: 0.6517Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6037 - acc: 0.6519 - val_loss: 6.8650 - val_acc: 0.5138 Epoch 4/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6209 - acc: 0.6506Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6045 - acc: 0.6516 - val_loss: 6.8767 - val_acc: 0.5162 Epoch 5/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6264 - acc: 0.6508Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6050 - acc: 0.6521 - val_loss: 6.8581 - val_acc: 0.5114 Epoch 6/50 6480/6680 [============================>.] - ETA: 0s - loss: 5.6172 - acc: 0.6509Epoch 00006: val_loss improved from 6.85458 to 6.83943, saving model to saved_models1/weights.best.vgg16_bs40_ep50.hdf5 6680/6680 [==============================] - 1s 160us/step - loss: 5.6034 - acc: 0.6518 - val_loss: 6.8394 - val_acc: 0.5126 Epoch 7/50 6560/6680 [============================>.] - ETA: 0s - loss: 5.5844 - acc: 0.6529Epoch 00007: val_loss improved from 6.83943 to 6.82535, saving model to saved_models1/weights.best.vgg16_bs40_ep50.hdf5 6680/6680 [==============================] - 1s 158us/step - loss: 5.6048 - acc: 0.6516 - val_loss: 6.8254 - val_acc: 0.5126 Epoch 8/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6092 - acc: 0.6512Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6052 - acc: 0.6515 - val_loss: 6.8258 - val_acc: 0.5126 Epoch 9/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6379 - acc: 0.6497Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6042 - acc: 0.6518 - val_loss: 6.8590 - val_acc: 0.5126 Epoch 10/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6187 - acc: 0.6509Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6047 - acc: 0.6518 - val_loss: 6.8386 - val_acc: 0.5126 Epoch 11/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.5939 - acc: 0.6526Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6046 - acc: 0.6519 - val_loss: 6.8385 - val_acc: 0.5126 Epoch 12/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6061 - acc: 0.6515Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6046 - acc: 0.6516 - val_loss: 6.8433 - val_acc: 0.5114 Epoch 13/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6045 - acc: 0.6518Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6054 - acc: 0.6518 - val_loss: 6.8543 - val_acc: 0.5090 Epoch 14/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6232 - acc: 0.6505Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6044 - acc: 0.6516 - val_loss: 6.8586 - val_acc: 0.5090 Epoch 15/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6104 - acc: 0.6512Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6039 - acc: 0.6516 - val_loss: 6.8651 - val_acc: 0.5054 Epoch 16/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6078 - acc: 0.6517Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6038 - acc: 0.6519 - val_loss: 6.8593 - val_acc: 0.5102 Epoch 17/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.5871 - acc: 0.6526Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6053 - acc: 0.6515 - val_loss: 6.8562 - val_acc: 0.5102 Epoch 18/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6005 - acc: 0.6518Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6039 - acc: 0.6516 - val_loss: 6.8449 - val_acc: 0.5102 Epoch 19/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6316 - acc: 0.6502Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6053 - acc: 0.6518 - val_loss: 6.8905 - val_acc: 0.5054 Epoch 20/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6235 - acc: 0.6506Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6046 - acc: 0.6518 - val_loss: 6.8910 - val_acc: 0.5090 Epoch 21/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6044 - acc: 0.6515Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 1s 159us/step - loss: 5.6053 - acc: 0.6515 - val_loss: 6.8793 - val_acc: 0.5114 Epoch 22/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6020 - acc: 0.6517Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6053 - acc: 0.6515 - val_loss: 6.8816 - val_acc: 0.5090 Epoch 23/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6137 - acc: 0.6512Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6047 - acc: 0.6518 - val_loss: 6.8727 - val_acc: 0.5102 Epoch 24/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6258 - acc: 0.6505Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6044 - acc: 0.6518 - val_loss: 6.8731 - val_acc: 0.5078 Epoch 25/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.5962 - acc: 0.6523Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6045 - acc: 0.6518 - val_loss: 6.8697 - val_acc: 0.5114 Epoch 26/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.5995 - acc: 0.6520Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6053 - acc: 0.6516 - val_loss: 6.8673 - val_acc: 0.5114 Epoch 27/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6119 - acc: 0.6512Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6053 - acc: 0.6516 - val_loss: 6.8905 - val_acc: 0.5090 Epoch 28/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6177 - acc: 0.6512Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6062 - acc: 0.6519 - val_loss: 6.8900 - val_acc: 0.5090 Epoch 29/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6087 - acc: 0.6515Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6047 - acc: 0.6518 - val_loss: 6.8958 - val_acc: 0.5138 Epoch 30/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6330 - acc: 0.6500Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6043 - acc: 0.6518 - val_loss: 6.8818 - val_acc: 0.5162 Epoch 31/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6088 - acc: 0.6515Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6048 - acc: 0.6518 - val_loss: 6.8841 - val_acc: 0.5126 Epoch 32/50 6480/6680 [============================>.] - ETA: 0s - loss: 5.6224 - acc: 0.6509Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 1s 159us/step - loss: 5.6037 - acc: 0.6521 - val_loss: 6.8863 - val_acc: 0.5078 Epoch 33/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.5968 - acc: 0.6523Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6051 - acc: 0.6518 - val_loss: 6.8655 - val_acc: 0.5078 Epoch 34/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6100 - acc: 0.6514Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6035 - acc: 0.6518 - val_loss: 6.8696 - val_acc: 0.5066 Epoch 35/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6053 - acc: 0.6515Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6038 - acc: 0.6516 - val_loss: 6.8501 - val_acc: 0.5090 Epoch 36/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.5902 - acc: 0.6529Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6059 - acc: 0.6519 - val_loss: 6.8472 - val_acc: 0.5138 Epoch 37/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.5881 - acc: 0.6531Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 1s 159us/step - loss: 5.6039 - acc: 0.6521 - val_loss: 6.8530 - val_acc: 0.5138 Epoch 38/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6080 - acc: 0.6517Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6039 - acc: 0.6519 - val_loss: 6.8452 - val_acc: 0.5102 Epoch 39/50 6560/6680 [============================>.] - ETA: 0s - loss: 5.5853 - acc: 0.6532Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6056 - acc: 0.6519 - val_loss: 6.8565 - val_acc: 0.5090 Epoch 40/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6272 - acc: 0.6505Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6052 - acc: 0.6518 - val_loss: 6.8572 - val_acc: 0.5114 Epoch 41/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.5898 - acc: 0.6528Epoch 00041: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6031 - acc: 0.6519 - val_loss: 6.8719 - val_acc: 0.5114 Epoch 42/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6176 - acc: 0.6509Epoch 00042: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6047 - acc: 0.6516 - val_loss: 6.8774 - val_acc: 0.5114 Epoch 43/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6000 - acc: 0.6521Epoch 00043: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6047 - acc: 0.6518 - val_loss: 6.8696 - val_acc: 0.5102 Epoch 44/50 6560/6680 [============================>.] - ETA: 0s - loss: 5.6065 - acc: 0.6517Epoch 00044: val_loss did not improve 6680/6680 [==============================] - 1s 156us/step - loss: 5.6047 - acc: 0.6518 - val_loss: 6.8733 - val_acc: 0.5090 Epoch 45/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6036 - acc: 0.6518Epoch 00045: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6045 - acc: 0.6518 - val_loss: 6.8639 - val_acc: 0.5150 Epoch 46/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.5922 - acc: 0.6525Epoch 00046: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6055 - acc: 0.6516 - val_loss: 6.8612 - val_acc: 0.5102 Epoch 47/50 6560/6680 [============================>.] - ETA: 0s - loss: 5.6219 - acc: 0.6506Epoch 00047: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6053 - acc: 0.6516 - val_loss: 6.8629 - val_acc: 0.5102 Epoch 48/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6229 - acc: 0.6508Epoch 00048: val_loss did not improve 6680/6680 [==============================] - 1s 158us/step - loss: 5.6041 - acc: 0.6519 - val_loss: 6.8646 - val_acc: 0.5090 Epoch 49/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.5914 - acc: 0.6526Epoch 00049: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6047 - acc: 0.6518 - val_loss: 6.8507 - val_acc: 0.5162 Epoch 50/50 6520/6680 [============================>.] - ETA: 0s - loss: 5.6135 - acc: 0.6512Epoch 00050: val_loss did not improve 6680/6680 [==============================] - 1s 157us/step - loss: 5.6045 - acc: 0.6518 - val_loss: 6.8720 - val_acc: 0.5138
In [26]:
import pandas as pd
pd.DataFrame(fitingdict_vgg16)
Out[26]:
| Batch_Size | Epochs | Test_Accuracy | |
|---|---|---|---|
| 0 | 20 | 20 | 45.813397 |
| 1 | 20 | 35 | 51.196172 |
| 2 | 20 | 40 | 52.033493 |
| 3 | 20 | 50 | 52.870813 |
| 4 | 35 | 20 | 52.631579 |
| 5 | 35 | 35 | 52.870813 |
| 6 | 35 | 40 | 53.110048 |
| 7 | 35 | 50 | 52.990431 |
| 8 | 37 | 20 | 52.990431 |
| 9 | 37 | 35 | 52.751196 |
| 10 | 37 | 40 | 52.631579 |
| 11 | 37 | 50 | 52.990431 |
| 12 | 40 | 20 | 52.511962 |
| 13 | 40 | 35 | 52.751196 |
| 14 | 40 | 40 | 52.272727 |
| 15 | 40 | 50 | 52.751196 |
Load the Model with the Best Validation Loss¶
In [21]:
#take largest testaccuracy's batch size and epochs
ind=fitingdict_vgg16['Test_Accuracy'].index(max(fitingdict_vgg16['Test_Accuracy']))
bs=fitingdict_vgg16['Batch_Size'][ind]
ep=fitingdict_vgg16['Epochs'][ind]
#LOAD the model with Best validation loss
VGG16_model.load_weights('saved_models1/weights.best.vgg16_bs'+str(bs)+'_ep'+str(ep)+'.hdf5')
Test the Model¶
Now, we can use the CNN to test how well it identifies breed within our test dataset of dog images. We print the test accuracy below.
In [20]:
# get index of predicted dog breed for each image in test set
VGG16_predictions = [np.argmax(VGG16_model.predict(np.expand_dims(feature, axis=0))) for feature in test_VGG16]
# report test accuracy
test_accuracy = 100*np.sum(np.array(VGG16_predictions)==np.argmax(test_targets, axis=1))/len(VGG16_predictions)
print('Test accuracy: %.4f%%' % test_accuracy)
Test accuracy: 53.1100%
Predict Dog Breed with the Model¶
In [23]:
from extract_bottleneck_features import *
def VGG16_predict_breed(img_path):
# extract bottleneck features
bottleneck_feature = extract_VGG16(path_to_tensor(img_path))
# obtain predicted vector
predicted_vector = VGG16_model.predict(bottleneck_feature)
# return dog breed that is predicted by the model
return dog_names[np.argmax(predicted_vector)]
In [24]:
### TODO: Write your algorithm.
### Feel free to use as many code cells as needed.
from keras.preprocessing import image
from os import walk
from os import listdir
from os.path import isfile, join
import random
import numpy as np
import cv2
def show_image(path):
img = image.load_img(path, target_size=(224, 224))
img = image.img_to_array(img)
plt.imshow(img/255)
plt.show()
def whos_face_is_this_VGG16(img_path):
if(dog_detector(img_path)):
print("\n**************************************")
show_image(img_path)
print("hello, Doggy!")
print("Your predicted breed is....")
print(VGG16_predict_breed(img_path))
elif(humanface_detector(img_path)):
print("\n**************************************")
show_image(img_path)
print("Hello, Human!")
print("You look like a.... ")
print(VGG16_predict_breed(img_path))
else:
print("\n**************************************")
show_image(img_path)
print("**No face detected..ERROR..**")
In [37]:
#Load the img files
imgs=["dog_images/doggy (1).jpg","dog_images/doggy (1a).jpg","Hooman_images/hooman (4).jpg","Hooman_images/hooman (10).jpg"]
for img in imgs:
whos_face_is_this_VGG16(img)
**************************************
hello, Doggy! Your predicted breed is.... Boykin_spaniel **************************************
hello, Doggy! Your predicted breed is.... Irish_water_spaniel **************************************
**No face detected..ERROR..** **************************************
Hello, Human! You look like a.... Dachshund
Above model VGG16 correctly differentiate 2 similar looking dogs into _Boykin_spaniel and Irish_waterspaniel . But, dog predicted _Boykinspaniel was actually _American_waterspaniel the wrong prediction here due to Test accuracy of the model is 53.1100% .
This was letter corrected by Xception model giving test acccuracy of 86.8421% .
Step 5: Create a CNN to Classify Dog Breeds (using Transfer Learning)¶
You will now use transfer learning to create a CNN that can identify dog breed from images. Your CNN must attain at least 60% accuracy on the test set.
In Step 4, we used transfer learning to create a CNN using VGG-16 bottleneck features. In this section, you must use the bottleneck features from a different pre-trained model. To make things easier for you, we have pre-computed the features for all of the networks that are currently available in Keras. These are already in the workspace, at /data/bottleneck_features. If you wish to download them on a different machine, they can be found at:
- VGG-19 bottleneck features
- ResNet-50 bottleneck features
- Inception bottleneck features
- Xception bottleneck features
The files are encoded as such:
Dog{network}Data.npz
where {network}, in the above filename, can be one of VGG19, Resnet50, InceptionV3, or Xception.
The above architectures are downloaded and stored for you in the /data/bottleneck_features/ folder.
This means the following will be in the /data/bottleneck_features/ folder:
DogVGG19Data.npz
DogResnet50Data.npz
DogInceptionV3Data.npz
DogXceptionData.npz
(IMPLEMENTATION) Obtain Bottleneck Features¶
In the code block below, extract the bottleneck features corresponding to the train, test, and validation sets by running the following:
bottleneck_features = np.load('/data/bottleneck_features/Dog{network}Data.npz')
train_{network} = bottleneck_features['train']
valid_{network} = bottleneck_features['valid']
test_{network} = bottleneck_features['test']Lests do with ResNet-50¶
In [32]:
### TODO: Obtain bottleneck features from another pre-trained CNN.
bottleneck_features = np.load('/data/bottleneck_features/DogResnet50Data.npz')
train_Resnet50 = bottleneck_features['train']
valid_Resnet50 = bottleneck_features['valid']
test_Resnet50 = bottleneck_features['test']
### TODO: Define your architecture.
from keras.layers import Conv2D, MaxPooling2D, GlobalAveragePooling2D
from keras.layers import Dropout, Flatten, Dense, Activation
from keras.models import Sequential
ResNet_model = Sequential()
ResNet_model.add(GlobalAveragePooling2D(input_shape=train_Resnet50.shape[1:]))
ResNet_model.add(Dense(133, activation='softmax'))
ResNet_model.summary()
_________________________________________________________________ Layer (type) Output Shape Param # ================================================================= global_average_pooling2d_4 ( (None, 2048) 0 _________________________________________________________________ dense_4 (Dense) (None, 133) 272517 ================================================================= Total params: 272,517 Trainable params: 272,517 Non-trainable params: 0 _________________________________________________________________
In [33]:
### TODO: Compile the model
ResNet_model.compile(loss="categorical_crossentropy",optimizer="rmsprop",metrics=["accuracy"])
### TODO: Train the model.
batch_size = [35,37,40,64]
epochs = [25,35,50]
fitingdict_ResNet={'Batch_Size':[],
'Epochs':[],
'Test_Accuracy':[]}
for bs in batch_size:
for ep in epochs:
checkpointer = ModelCheckpoint(filepath='saved_models3/weights.best.ResNet_bs'+str(bs)+'_ep'+str(ep)+'.hdf5',
verbose=1, save_best_only=True)
print("\nBatch size={0} Epoch={1}".format(bs,ep))
ResNet_model.fit(train_Resnet50, train_targets,validation_data=(valid_Resnet50, valid_targets),
epochs=ep , batch_size=bs,
callbacks=[checkpointer],verbose=1)
#LOAD the model with Best validation loss
ResNet_model.load_weights('saved_models3/weights.best.ResNet_bs'+str(bs)+'_ep'+str(ep)+'.hdf5')
ResNet_predictions = [np.argmax(ResNet_model.predict(np.expand_dims(feature, axis=0))) for feature in test_Resnet50]
test_accuracy = 100*np.sum(np.array(ResNet_predictions)==np.argmax(test_targets, axis=1))/len(ResNet_predictions)
fitingdict_ResNet['Batch_Size'].append(bs)
fitingdict_ResNet['Epochs'].append(ep)
fitingdict_ResNet['Test_Accuracy'].append(test_accuracy)
Batch size=35 Epoch=25 Train on 6680 samples, validate on 835 samples Epoch 1/25 6545/6680 [============================>.] - ETA: 0s - loss: 1.8384 - acc: 0.5630Epoch 00001: val_loss improved from inf to 0.89360, saving model to saved_models3/weights.best.ResNet_bs35_ep25.hdf5 6680/6680 [==============================] - 1s 204us/step - loss: 1.8207 - acc: 0.5659 - val_loss: 0.8936 - val_acc: 0.7246 Epoch 2/25 6475/6680 [============================>.] - ETA: 0s - loss: 0.4679 - acc: 0.8593Epoch 00002: val_loss improved from 0.89360 to 0.72643, saving model to saved_models3/weights.best.ResNet_bs35_ep25.hdf5 6680/6680 [==============================] - 1s 152us/step - loss: 0.4684 - acc: 0.8594 - val_loss: 0.7264 - val_acc: 0.7808 Epoch 3/25 6370/6680 [===========================>..] - ETA: 0s - loss: 0.2605 - acc: 0.9226Epoch 00003: val_loss improved from 0.72643 to 0.64135, saving model to saved_models3/weights.best.ResNet_bs35_ep25.hdf5 6680/6680 [==============================] - 1s 145us/step - loss: 0.2609 - acc: 0.9213 - val_loss: 0.6414 - val_acc: 0.7988 Epoch 4/25 6580/6680 [============================>.] - ETA: 0s - loss: 0.1601 - acc: 0.9541Epoch 00004: val_loss improved from 0.64135 to 0.62651, saving model to saved_models3/weights.best.ResNet_bs35_ep25.hdf5 6680/6680 [==============================] - 1s 140us/step - loss: 0.1594 - acc: 0.9543 - val_loss: 0.6265 - val_acc: 0.8144 Epoch 5/25 6475/6680 [============================>.] - ETA: 0s - loss: 0.1107 - acc: 0.9705Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.1129 - acc: 0.9696 - val_loss: 0.6381 - val_acc: 0.8096 Epoch 6/25 6440/6680 [===========================>..] - ETA: 0s - loss: 0.0738 - acc: 0.9825Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0743 - acc: 0.9823 - val_loss: 0.6644 - val_acc: 0.8144 Epoch 7/25 6405/6680 [===========================>..] - ETA: 0s - loss: 0.0509 - acc: 0.9880Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0516 - acc: 0.9879 - val_loss: 0.6700 - val_acc: 0.8084 Epoch 8/25 6580/6680 [============================>.] - ETA: 0s - loss: 0.0367 - acc: 0.9910Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0367 - acc: 0.9910 - val_loss: 0.6378 - val_acc: 0.8192 Epoch 9/25 6580/6680 [============================>.] - ETA: 0s - loss: 0.0266 - acc: 0.9935Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0266 - acc: 0.9936 - val_loss: 0.6611 - val_acc: 0.8120 Epoch 10/25 6545/6680 [============================>.] - ETA: 0s - loss: 0.0197 - acc: 0.9960Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0197 - acc: 0.9961 - val_loss: 0.6787 - val_acc: 0.8275 Epoch 11/25 6580/6680 [============================>.] - ETA: 0s - loss: 0.0161 - acc: 0.9970Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0159 - acc: 0.9970 - val_loss: 0.6935 - val_acc: 0.8275 Epoch 12/25 6440/6680 [===========================>..] - ETA: 0s - loss: 0.0118 - acc: 0.9974Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 141us/step - loss: 0.0120 - acc: 0.9973 - val_loss: 0.6982 - val_acc: 0.8371 Epoch 13/25 6475/6680 [============================>.] - ETA: 0s - loss: 0.0118 - acc: 0.9981Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0117 - acc: 0.9981 - val_loss: 0.7255 - val_acc: 0.8275 Epoch 14/25 6580/6680 [============================>.] - ETA: 0s - loss: 0.0085 - acc: 0.9980Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0084 - acc: 0.9981 - val_loss: 0.7602 - val_acc: 0.8371 Epoch 15/25 6580/6680 [============================>.] - ETA: 0s - loss: 0.0074 - acc: 0.9982Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0073 - acc: 0.9982 - val_loss: 0.8233 - val_acc: 0.8132 Epoch 16/25 6405/6680 [===========================>..] - ETA: 0s - loss: 0.0066 - acc: 0.9986Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0067 - acc: 0.9985 - val_loss: 0.8173 - val_acc: 0.8263 Epoch 17/25 6580/6680 [============================>.] - ETA: 0s - loss: 0.0060 - acc: 0.9982Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0060 - acc: 0.9982 - val_loss: 0.8164 - val_acc: 0.8299 Epoch 18/25 6475/6680 [============================>.] - ETA: 0s - loss: 0.0046 - acc: 0.9985Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0045 - acc: 0.9985 - val_loss: 0.7978 - val_acc: 0.8347 Epoch 19/25 6510/6680 [============================>.] - ETA: 0s - loss: 0.0060 - acc: 0.9986Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0058 - acc: 0.9987 - val_loss: 0.8421 - val_acc: 0.8407 Epoch 20/25 6370/6680 [===========================>..] - ETA: 0s - loss: 0.0051 - acc: 0.9986Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0049 - acc: 0.9987 - val_loss: 0.8878 - val_acc: 0.8323 Epoch 21/25 6335/6680 [===========================>..] - ETA: 0s - loss: 0.0032 - acc: 0.9984Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0031 - acc: 0.9985 - val_loss: 0.9068 - val_acc: 0.8299 Epoch 22/25 6545/6680 [============================>.] - ETA: 0s - loss: 0.0049 - acc: 0.9986Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0048 - acc: 0.9987 - val_loss: 0.8950 - val_acc: 0.8383 Epoch 23/25 6440/6680 [===========================>..] - ETA: 0s - loss: 0.0032 - acc: 0.9988Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0039 - acc: 0.9987 - val_loss: 0.9336 - val_acc: 0.8407 Epoch 24/25 6475/6680 [============================>.] - ETA: 0s - loss: 0.0043 - acc: 0.9991Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0042 - acc: 0.9991 - val_loss: 0.9333 - val_acc: 0.8263 Epoch 25/25 6265/6680 [===========================>..] - ETA: 0s - loss: 0.0041 - acc: 0.9986Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 1s 138us/step - loss: 0.0040 - acc: 0.9985 - val_loss: 0.9509 - val_acc: 0.8371 Batch size=35 Epoch=35 Train on 6680 samples, validate on 835 samples Epoch 1/35 6615/6680 [============================>.] - ETA: 0s - loss: 0.1190 - acc: 0.9664Epoch 00001: val_loss improved from inf to 0.64755, saving model to saved_models3/weights.best.ResNet_bs35_ep35.hdf5 6680/6680 [==============================] - 1s 140us/step - loss: 0.1187 - acc: 0.9666 - val_loss: 0.6475 - val_acc: 0.7976 Epoch 2/35 6580/6680 [============================>.] - ETA: 0s - loss: 0.0740 - acc: 0.9796Epoch 00002: val_loss improved from 0.64755 to 0.61231, saving model to saved_models3/weights.best.ResNet_bs35_ep35.hdf5 6680/6680 [==============================] - 1s 142us/step - loss: 0.0746 - acc: 0.9793 - val_loss: 0.6123 - val_acc: 0.8120 Epoch 3/35 6545/6680 [============================>.] - ETA: 0s - loss: 0.0515 - acc: 0.9859Epoch 00003: val_loss improved from 0.61231 to 0.60656, saving model to saved_models3/weights.best.ResNet_bs35_ep35.hdf5 6680/6680 [==============================] - 1s 141us/step - loss: 0.0514 - acc: 0.9861 - val_loss: 0.6066 - val_acc: 0.8419 Epoch 4/35 6615/6680 [============================>.] - ETA: 0s - loss: 0.0375 - acc: 0.9908Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 1s 138us/step - loss: 0.0375 - acc: 0.9907 - val_loss: 0.6700 - val_acc: 0.8251 Epoch 5/35 6650/6680 [============================>.] - ETA: 0s - loss: 0.0271 - acc: 0.9938Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 138us/step - loss: 0.0273 - acc: 0.9937 - val_loss: 0.7036 - val_acc: 0.8251 Epoch 6/35 6545/6680 [============================>.] - ETA: 0s - loss: 0.0206 - acc: 0.9954Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0204 - acc: 0.9955 - val_loss: 0.6562 - val_acc: 0.8299 Epoch 7/35 6615/6680 [============================>.] - ETA: 0s - loss: 0.0160 - acc: 0.9979Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0160 - acc: 0.9979 - val_loss: 0.7027 - val_acc: 0.8240 Epoch 8/35 6650/6680 [============================>.] - ETA: 0s - loss: 0.0128 - acc: 0.9973Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 138us/step - loss: 0.0128 - acc: 0.9973 - val_loss: 0.7094 - val_acc: 0.8251 Epoch 9/35 6615/6680 [============================>.] - ETA: 0s - loss: 0.0091 - acc: 0.9983Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 138us/step - loss: 0.0091 - acc: 0.9984 - val_loss: 0.7685 - val_acc: 0.8275 Epoch 10/35 6265/6680 [===========================>..] - ETA: 0s - loss: 0.0077 - acc: 0.9981Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 138us/step - loss: 0.0075 - acc: 0.9982 - val_loss: 0.7474 - val_acc: 0.8335 Epoch 11/35 6440/6680 [===========================>..] - ETA: 0s - loss: 0.0067 - acc: 0.9988Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0068 - acc: 0.9987 - val_loss: 0.7975 - val_acc: 0.8299 Epoch 12/35 6440/6680 [===========================>..] - ETA: 0s - loss: 0.0044 - acc: 0.9992Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0062 - acc: 0.9987 - val_loss: 0.8197 - val_acc: 0.8347 Epoch 13/35 6545/6680 [============================>.] - ETA: 0s - loss: 0.0058 - acc: 0.9983Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0057 - acc: 0.9984 - val_loss: 0.8242 - val_acc: 0.8228 Epoch 14/35 6370/6680 [===========================>..] - ETA: 0s - loss: 0.0037 - acc: 0.9991Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 141us/step - loss: 0.0038 - acc: 0.9990 - val_loss: 0.8311 - val_acc: 0.8323 Epoch 15/35 6545/6680 [============================>.] - ETA: 0s - loss: 0.0049 - acc: 0.9985Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0048 - acc: 0.9985 - val_loss: 0.8865 - val_acc: 0.8240 Epoch 16/35 6370/6680 [===========================>..] - ETA: 0s - loss: 0.0048 - acc: 0.9989Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 141us/step - loss: 0.0047 - acc: 0.9990 - val_loss: 0.8887 - val_acc: 0.8347 Epoch 17/35 6440/6680 [===========================>..] - ETA: 0s - loss: 0.0053 - acc: 0.9984Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 141us/step - loss: 0.0052 - acc: 0.9985 - val_loss: 0.9248 - val_acc: 0.8299 Epoch 18/35 6405/6680 [===========================>..] - ETA: 0s - loss: 0.0046 - acc: 0.9986Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0045 - acc: 0.9987 - val_loss: 0.9275 - val_acc: 0.8251 Epoch 19/35 6370/6680 [===========================>..] - ETA: 0s - loss: 0.0035 - acc: 0.9989Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 141us/step - loss: 0.0051 - acc: 0.9988 - val_loss: 0.8996 - val_acc: 0.8275 Epoch 20/35 6510/6680 [============================>.] - ETA: 0s - loss: 0.0043 - acc: 0.9985Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0043 - acc: 0.9985 - val_loss: 0.9309 - val_acc: 0.8299 Epoch 21/35 6370/6680 [===========================>..] - ETA: 0s - loss: 0.0047 - acc: 0.9987Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0045 - acc: 0.9988 - val_loss: 0.9384 - val_acc: 0.8240 Epoch 22/35 6580/6680 [============================>.] - ETA: 0s - loss: 0.0045 - acc: 0.9986Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0045 - acc: 0.9987 - val_loss: 0.9500 - val_acc: 0.8240 Epoch 23/35 6370/6680 [===========================>..] - ETA: 0s - loss: 0.0045 - acc: 0.9986Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0043 - acc: 0.9987 - val_loss: 0.9678 - val_acc: 0.8263 Epoch 24/35 6545/6680 [============================>.] - ETA: 0s - loss: 0.0036 - acc: 0.9986Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0035 - acc: 0.9987 - val_loss: 0.9874 - val_acc: 0.8263 Epoch 25/35 6440/6680 [===========================>..] - ETA: 0s - loss: 0.0046 - acc: 0.9984Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0045 - acc: 0.9985 - val_loss: 1.0458 - val_acc: 0.8132 Epoch 26/35 6580/6680 [============================>.] - ETA: 0s - loss: 0.0040 - acc: 0.9988Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0039 - acc: 0.9988 - val_loss: 1.0190 - val_acc: 0.8228 Epoch 27/35 6475/6680 [============================>.] - ETA: 0s - loss: 0.0036 - acc: 0.9988Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0039 - acc: 0.9987 - val_loss: 1.0086 - val_acc: 0.8299 Epoch 28/35 6545/6680 [============================>.] - ETA: 0s - loss: 0.0043 - acc: 0.9985Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0042 - acc: 0.9985 - val_loss: 1.0147 - val_acc: 0.8287 Epoch 29/35 6440/6680 [===========================>..] - ETA: 0s - loss: 0.0048 - acc: 0.9986Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0048 - acc: 0.9985 - val_loss: 1.0259 - val_acc: 0.8311 Epoch 30/35 6580/6680 [============================>.] - ETA: 0s - loss: 0.0041 - acc: 0.9986Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0040 - acc: 0.9987 - val_loss: 1.0707 - val_acc: 0.8395 Epoch 31/35 6440/6680 [===========================>..] - ETA: 0s - loss: 0.0040 - acc: 0.9984Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0038 - acc: 0.9985 - val_loss: 1.0625 - val_acc: 0.8347 Epoch 32/35 6510/6680 [============================>.] - ETA: 0s - loss: 0.0040 - acc: 0.9989Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0045 - acc: 0.9988 - val_loss: 1.1262 - val_acc: 0.8228 Epoch 33/35 6510/6680 [============================>.] - ETA: 0s - loss: 0.0051 - acc: 0.9985Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0050 - acc: 0.9985 - val_loss: 1.0909 - val_acc: 0.8347 Epoch 34/35 6580/6680 [============================>.] - ETA: 0s - loss: 0.0045 - acc: 0.9988 Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0044 - acc: 0.9988 - val_loss: 1.0839 - val_acc: 0.8395 Epoch 35/35 6580/6680 [============================>.] - ETA: 0s - loss: 0.0040 - acc: 0.9986Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0040 - acc: 0.9987 - val_loss: 1.1221 - val_acc: 0.8335 Batch size=35 Epoch=50 Train on 6680 samples, validate on 835 samples Epoch 1/50 6580/6680 [============================>.] - ETA: 0s - loss: 0.0412 - acc: 0.9900Epoch 00001: val_loss improved from inf to 0.63494, saving model to saved_models3/weights.best.ResNet_bs35_ep50.hdf5 6680/6680 [==============================] - 1s 141us/step - loss: 0.0416 - acc: 0.9900 - val_loss: 0.6349 - val_acc: 0.8287 Epoch 2/50 6475/6680 [============================>.] - ETA: 0s - loss: 0.0271 - acc: 0.9943Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0276 - acc: 0.9942 - val_loss: 0.6757 - val_acc: 0.8299 Epoch 3/50 6475/6680 [============================>.] - ETA: 0s - loss: 0.0206 - acc: 0.9968Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0205 - acc: 0.9969 - val_loss: 0.7123 - val_acc: 0.8108 Epoch 4/50 6475/6680 [============================>.] - ETA: 0s - loss: 0.0153 - acc: 0.9964Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0152 - acc: 0.9964 - val_loss: 0.7267 - val_acc: 0.8204 Epoch 5/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.0124 - acc: 0.9983Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 138us/step - loss: 0.0125 - acc: 0.9984 - val_loss: 0.7871 - val_acc: 0.8168 Epoch 6/50 6510/6680 [============================>.] - ETA: 0s - loss: 0.0101 - acc: 0.9978Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0101 - acc: 0.9978 - val_loss: 0.7250 - val_acc: 0.8228 Epoch 7/50 6545/6680 [============================>.] - ETA: 0s - loss: 0.0076 - acc: 0.9989Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0077 - acc: 0.9988 - val_loss: 0.7569 - val_acc: 0.8311 Epoch 8/50 6510/6680 [============================>.] - ETA: 0s - loss: 0.0082 - acc: 0.9983Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0081 - acc: 0.9984 - val_loss: 0.7588 - val_acc: 0.8443 Epoch 9/50 6440/6680 [===========================>..] - ETA: 0s - loss: 0.0073 - acc: 0.9981Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0072 - acc: 0.9982 - val_loss: 0.8089 - val_acc: 0.8287 Epoch 10/50 6545/6680 [============================>.] - ETA: 0s - loss: 0.0062 - acc: 0.9988Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0061 - acc: 0.9988 - val_loss: 0.7965 - val_acc: 0.8275 Epoch 11/50 6510/6680 [============================>.] - ETA: 0s - loss: 0.0049 - acc: 0.9986Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0054 - acc: 0.9985 - val_loss: 0.8030 - val_acc: 0.8347 Epoch 12/50 6510/6680 [============================>.] - ETA: 0s - loss: 0.0049 - acc: 0.9983Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0047 - acc: 0.9984 - val_loss: 0.8322 - val_acc: 0.8287 Epoch 13/50 6300/6680 [===========================>..] - ETA: 0s - loss: 0.0062 - acc: 0.9989Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 138us/step - loss: 0.0059 - acc: 0.9990 - val_loss: 0.8411 - val_acc: 0.8323 Epoch 14/50 6545/6680 [============================>.] - ETA: 0s - loss: 0.0053 - acc: 0.9988Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0052 - acc: 0.9988 - val_loss: 0.8733 - val_acc: 0.8335 Epoch 15/50 6545/6680 [============================>.] - ETA: 0s - loss: 0.0044 - acc: 0.9985Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0044 - acc: 0.9985 - val_loss: 0.9026 - val_acc: 0.8311 Epoch 16/50 6510/6680 [============================>.] - ETA: 0s - loss: 0.0049 - acc: 0.9989Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0048 - acc: 0.9990 - val_loss: 0.9034 - val_acc: 0.8359 Epoch 17/50 6405/6680 [===========================>..] - ETA: 0s - loss: 0.0053 - acc: 0.9986Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0051 - acc: 0.9987 - val_loss: 0.9373 - val_acc: 0.8251 Epoch 18/50 6475/6680 [============================>.] - ETA: 0s - loss: 0.0054 - acc: 0.9988Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0052 - acc: 0.9988 - val_loss: 0.9476 - val_acc: 0.8311 Epoch 19/50 6265/6680 [===========================>..] - ETA: 0s - loss: 0.0058 - acc: 0.9987Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 138us/step - loss: 0.0054 - acc: 0.9988 - val_loss: 0.9621 - val_acc: 0.8311 Epoch 20/50 6545/6680 [============================>.] - ETA: 0s - loss: 0.0054 - acc: 0.9986Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0053 - acc: 0.9987 - val_loss: 0.9760 - val_acc: 0.8251 Epoch 21/50 6475/6680 [============================>.] - ETA: 0s - loss: 0.0041 - acc: 0.9988Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0046 - acc: 0.9987 - val_loss: 0.9898 - val_acc: 0.8275 Epoch 22/50 6475/6680 [============================>.] - ETA: 0s - loss: 0.0032 - acc: 0.9989Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0037 - acc: 0.9988 - val_loss: 0.9913 - val_acc: 0.8275 Epoch 23/50 6440/6680 [===========================>..] - ETA: 0s - loss: 0.0040 - acc: 0.9988Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0039 - acc: 0.9988 - val_loss: 1.0225 - val_acc: 0.8192 Epoch 24/50 6510/6680 [============================>.] - ETA: 0s - loss: 0.0044 - acc: 0.9986Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0042 - acc: 0.9987 - val_loss: 1.0019 - val_acc: 0.8251 Epoch 25/50 6615/6680 [============================>.] - ETA: 0s - loss: 0.0041 - acc: 0.9989Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0043 - acc: 0.9988 - val_loss: 1.0473 - val_acc: 0.8383 Epoch 26/50 6580/6680 [============================>.] - ETA: 0s - loss: 0.0031 - acc: 0.9988Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0037 - acc: 0.9987 - val_loss: 1.0357 - val_acc: 0.8311 Epoch 27/50 6615/6680 [============================>.] - ETA: 0s - loss: 0.0043 - acc: 0.9989Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 1s 138us/step - loss: 0.0042 - acc: 0.9990 - val_loss: 1.0419 - val_acc: 0.8299 Epoch 28/50 6615/6680 [============================>.] - ETA: 0s - loss: 0.0035 - acc: 0.9989Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0034 - acc: 0.9990 - val_loss: 1.1074 - val_acc: 0.8371 Epoch 29/50 6440/6680 [===========================>..] - ETA: 0s - loss: 0.0044 - acc: 0.9986Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0046 - acc: 0.9985 - val_loss: 1.0703 - val_acc: 0.8251 Epoch 30/50 6475/6680 [============================>.] - ETA: 0s - loss: 0.0051 - acc: 0.9988Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0050 - acc: 0.9988 - val_loss: 1.0758 - val_acc: 0.8287 Epoch 31/50 6265/6680 [===========================>..] - ETA: 0s - loss: 0.0035 - acc: 0.9986Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 1s 138us/step - loss: 0.0038 - acc: 0.9985 - val_loss: 1.1536 - val_acc: 0.8228 Epoch 32/50 6510/6680 [============================>.] - ETA: 0s - loss: 0.0032 - acc: 0.9991Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0043 - acc: 0.9990 - val_loss: 1.0881 - val_acc: 0.8371 Epoch 33/50 6475/6680 [============================>.] - ETA: 0s - loss: 0.0033 - acc: 0.9989Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0033 - acc: 0.9990 - val_loss: 1.1016 - val_acc: 0.8299 Epoch 34/50 6545/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9986Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 1s 148us/step - loss: 0.0049 - acc: 0.9985 - val_loss: 1.1210 - val_acc: 0.8359 Epoch 35/50 6475/6680 [============================>.] - ETA: 0s - loss: 0.0045 - acc: 0.9988Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 1s 141us/step - loss: 0.0052 - acc: 0.9987 - val_loss: 1.1178 - val_acc: 0.8299 Epoch 36/50 6405/6680 [===========================>..] - ETA: 0s - loss: 0.0040 - acc: 0.9989Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0038 - acc: 0.9990 - val_loss: 1.1435 - val_acc: 0.8311 Epoch 37/50 6405/6680 [===========================>..] - ETA: 0s - loss: 0.0033 - acc: 0.9989Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0040 - acc: 0.9988 - val_loss: 1.1200 - val_acc: 0.8251 Epoch 38/50 6475/6680 [============================>.] - ETA: 0s - loss: 0.0047 - acc: 0.9986Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0046 - acc: 0.9987 - val_loss: 1.1427 - val_acc: 0.8323 Epoch 39/50 6580/6680 [============================>.] - ETA: 0s - loss: 0.0037 - acc: 0.9989Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0043 - acc: 0.9988 - val_loss: 1.1291 - val_acc: 0.8383 Epoch 40/50 6545/6680 [============================>.] - ETA: 0s - loss: 0.0050 - acc: 0.9986Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0049 - acc: 0.9987 - val_loss: 1.1254 - val_acc: 0.8371 Epoch 41/50 6440/6680 [===========================>..] - ETA: 0s - loss: 0.0042 - acc: 0.9991Epoch 00041: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0040 - acc: 0.9991 - val_loss: 1.1342 - val_acc: 0.8359 Epoch 42/50 6545/6680 [============================>.] - ETA: 0s - loss: 0.0040 - acc: 0.9989Epoch 00042: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0042 - acc: 0.9988 - val_loss: 1.1426 - val_acc: 0.8275 Epoch 43/50 6510/6680 [============================>.] - ETA: 0s - loss: 0.0045 - acc: 0.9985Epoch 00043: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0044 - acc: 0.9985 - val_loss: 1.1492 - val_acc: 0.8347 Epoch 44/50 6510/6680 [============================>.] - ETA: 0s - loss: 0.0052 - acc: 0.9988Epoch 00044: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0050 - acc: 0.9988 - val_loss: 1.1341 - val_acc: 0.8371 Epoch 45/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.0045 - acc: 0.9988Epoch 00045: val_loss did not improve 6680/6680 [==============================] - 1s 139us/step - loss: 0.0045 - acc: 0.9988 - val_loss: 1.1608 - val_acc: 0.8383 Epoch 46/50 6440/6680 [===========================>..] - ETA: 0s - loss: 0.0051 - acc: 0.9988Epoch 00046: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0049 - acc: 0.9988 - val_loss: 1.1535 - val_acc: 0.8335 Epoch 47/50 6405/6680 [===========================>..] - ETA: 0s - loss: 0.0033 - acc: 0.9986Epoch 00047: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0039 - acc: 0.9985 - val_loss: 1.1879 - val_acc: 0.8347 Epoch 48/50 6475/6680 [============================>.] - ETA: 0s - loss: 0.0045 - acc: 0.9986Epoch 00048: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0043 - acc: 0.9987 - val_loss: 1.1485 - val_acc: 0.8359 Epoch 49/50 6440/6680 [===========================>..] - ETA: 0s - loss: 0.0047 - acc: 0.9983Epoch 00049: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0045 - acc: 0.9984 - val_loss: 1.1493 - val_acc: 0.8335 Epoch 50/50 6510/6680 [============================>.] - ETA: 0s - loss: 0.0039 - acc: 0.9988Epoch 00050: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0044 - acc: 0.9987 - val_loss: 1.1813 - val_acc: 0.8371 Batch size=37 Epoch=25 Train on 6680 samples, validate on 835 samples Epoch 1/25 6290/6680 [===========================>..] - ETA: 0s - loss: 0.0326 - acc: 0.9924Epoch 00001: val_loss improved from inf to 0.68743, saving model to saved_models3/weights.best.ResNet_bs37_ep25.hdf5 6680/6680 [==============================] - 1s 135us/step - loss: 0.0331 - acc: 0.9925 - val_loss: 0.6874 - val_acc: 0.8275 Epoch 2/25 6290/6680 [===========================>..] - ETA: 0s - loss: 0.0213 - acc: 0.9952Epoch 00002: val_loss improved from 0.68743 to 0.67728, saving model to saved_models3/weights.best.ResNet_bs37_ep25.hdf5 6680/6680 [==============================] - 1s 135us/step - loss: 0.0209 - acc: 0.9954 - val_loss: 0.6773 - val_acc: 0.8168 Epoch 3/25 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0164 - acc: 0.9973Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0161 - acc: 0.9975 - val_loss: 0.6920 - val_acc: 0.8479 Epoch 4/25 6364/6680 [===========================>..] - ETA: 0s - loss: 0.0130 - acc: 0.9969Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0126 - acc: 0.9970 - val_loss: 0.7347 - val_acc: 0.8144 Epoch 5/25 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0086 - acc: 0.9983Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0090 - acc: 0.9982 - val_loss: 0.7600 - val_acc: 0.8120 Epoch 6/25 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0074 - acc: 0.9984Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0076 - acc: 0.9982 - val_loss: 0.7362 - val_acc: 0.8275 Epoch 7/25 6401/6680 [===========================>..] - ETA: 0s - loss: 0.0074 - acc: 0.9981Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0073 - acc: 0.9982 - val_loss: 0.7607 - val_acc: 0.8347 Epoch 8/25 6364/6680 [===========================>..] - ETA: 0s - loss: 0.0058 - acc: 0.9986Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0058 - acc: 0.9985 - val_loss: 0.7961 - val_acc: 0.8275 Epoch 9/25 6660/6680 [============================>.] - ETA: 0s - loss: 0.0054 - acc: 0.9988Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0054 - acc: 0.9988 - val_loss: 0.8090 - val_acc: 0.8216 Epoch 10/25 6364/6680 [===========================>..] - ETA: 0s - loss: 0.0062 - acc: 0.9980Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0060 - acc: 0.9981 - val_loss: 0.8194 - val_acc: 0.8216 Epoch 11/25 6438/6680 [===========================>..] - ETA: 0s - loss: 0.0049 - acc: 0.9988Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0050 - acc: 0.9987 - val_loss: 0.8556 - val_acc: 0.8299 Epoch 12/25 6660/6680 [============================>.] - ETA: 0s - loss: 0.0055 - acc: 0.9989Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0055 - acc: 0.9990 - val_loss: 0.8532 - val_acc: 0.8347 Epoch 13/25 6660/6680 [============================>.] - ETA: 0s - loss: 0.0047 - acc: 0.9982Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0047 - acc: 0.9982 - val_loss: 0.8657 - val_acc: 0.8323 Epoch 14/25 6660/6680 [============================>.] - ETA: 0s - loss: 0.0039 - acc: 0.9983Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0039 - acc: 0.9984 - val_loss: 0.8786 - val_acc: 0.8228 Epoch 15/25 6623/6680 [============================>.] - ETA: 0s - loss: 0.0056 - acc: 0.9985Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0056 - acc: 0.9985 - val_loss: 0.9357 - val_acc: 0.8204 Epoch 16/25 6253/6680 [===========================>..] - ETA: 0s - loss: 0.0033 - acc: 0.9989Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0037 - acc: 0.9987 - val_loss: 0.9410 - val_acc: 0.8335 Epoch 17/25 6512/6680 [============================>.] - ETA: 0s - loss: 0.0029 - acc: 0.9988Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 141us/step - loss: 0.0028 - acc: 0.9988 - val_loss: 0.9484 - val_acc: 0.8240 Epoch 18/25 6549/6680 [============================>.] - ETA: 0s - loss: 0.0040 - acc: 0.9989Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 142us/step - loss: 0.0041 - acc: 0.9988 - val_loss: 0.9807 - val_acc: 0.8311 Epoch 19/25 6549/6680 [============================>.] - ETA: 0s - loss: 0.0027 - acc: 0.9988Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 142us/step - loss: 0.0031 - acc: 0.9987 - val_loss: 0.9920 - val_acc: 0.8228 Epoch 20/25 6512/6680 [============================>.] - ETA: 0s - loss: 0.0050 - acc: 0.9986Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 142us/step - loss: 0.0049 - acc: 0.9987 - val_loss: 1.0131 - val_acc: 0.8156 Epoch 21/25 6549/6680 [============================>.] - ETA: 0s - loss: 0.0041 - acc: 0.9985Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 1s 142us/step - loss: 0.0040 - acc: 0.9985 - val_loss: 1.0180 - val_acc: 0.8240 Epoch 22/25 6549/6680 [============================>.] - ETA: 0s - loss: 0.0047 - acc: 0.9986Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 1s 142us/step - loss: 0.0046 - acc: 0.9987 - val_loss: 0.9996 - val_acc: 0.8251 Epoch 23/25 6512/6680 [============================>.] - ETA: 0s - loss: 0.0044 - acc: 0.9986Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 1s 141us/step - loss: 0.0043 - acc: 0.9987 - val_loss: 1.0166 - val_acc: 0.8323 Epoch 24/25 6512/6680 [============================>.] - ETA: 0s - loss: 0.0045 - acc: 0.9983Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 1s 141us/step - loss: 0.0044 - acc: 0.9984 - val_loss: 1.0523 - val_acc: 0.8287 Epoch 25/25 6512/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9989Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 1s 142us/step - loss: 0.0041 - acc: 0.9990 - val_loss: 1.0293 - val_acc: 0.8299 Batch size=37 Epoch=35 Train on 6680 samples, validate on 835 samples Epoch 1/35 6549/6680 [============================>.] - ETA: 0s - loss: 0.0198 - acc: 0.9953Epoch 00001: val_loss improved from inf to 0.73685, saving model to saved_models3/weights.best.ResNet_bs37_ep35.hdf5 6680/6680 [==============================] - 1s 141us/step - loss: 0.0201 - acc: 0.9952 - val_loss: 0.7368 - val_acc: 0.8072 Epoch 2/35 6549/6680 [============================>.] - ETA: 0s - loss: 0.0120 - acc: 0.9965Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0119 - acc: 0.9966 - val_loss: 0.7697 - val_acc: 0.8251 Epoch 3/35 6549/6680 [============================>.] - ETA: 0s - loss: 0.0105 - acc: 0.9974Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 1s 141us/step - loss: 0.0103 - acc: 0.9975 - val_loss: 0.7390 - val_acc: 0.8263 Epoch 4/35 6549/6680 [============================>.] - ETA: 0s - loss: 0.0089 - acc: 0.9986Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0090 - acc: 0.9985 - val_loss: 0.7394 - val_acc: 0.8240 Epoch 5/35 6549/6680 [============================>.] - ETA: 0s - loss: 0.0066 - acc: 0.9983Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 140us/step - loss: 0.0075 - acc: 0.9981 - val_loss: 0.7927 - val_acc: 0.8228 Epoch 6/35 6512/6680 [============================>.] - ETA: 0s - loss: 0.0067 - acc: 0.9988Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 141us/step - loss: 0.0069 - acc: 0.9985 - val_loss: 0.8260 - val_acc: 0.8263 Epoch 7/35 6364/6680 [===========================>..] - ETA: 0s - loss: 0.0054 - acc: 0.9984Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0061 - acc: 0.9982 - val_loss: 0.8256 - val_acc: 0.8216 Epoch 8/35 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0054 - acc: 0.9987Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0054 - acc: 0.9985 - val_loss: 0.8441 - val_acc: 0.8263 Epoch 9/35 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0045 - acc: 0.9989Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0048 - acc: 0.9988 - val_loss: 0.8894 - val_acc: 0.8240 Epoch 10/35 6438/6680 [===========================>..] - ETA: 0s - loss: 0.0044 - acc: 0.9984Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0043 - acc: 0.9985 - val_loss: 0.7931 - val_acc: 0.8335 Epoch 11/35 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0039 - acc: 0.9986Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0038 - acc: 0.9987 - val_loss: 0.8817 - val_acc: 0.8347 Epoch 12/35 6290/6680 [===========================>..] - ETA: 0s - loss: 0.0039 - acc: 0.9989Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0039 - acc: 0.9988 - val_loss: 0.9138 - val_acc: 0.8144 Epoch 13/35 6290/6680 [===========================>..] - ETA: 0s - loss: 0.0050 - acc: 0.9987Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0047 - acc: 0.9988 - val_loss: 0.9537 - val_acc: 0.8287 Epoch 14/35 6623/6680 [============================>.] - ETA: 0s - loss: 0.0041 - acc: 0.9989Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0040 - acc: 0.9990 - val_loss: 0.9308 - val_acc: 0.8251 Epoch 15/35 6401/6680 [===========================>..] - ETA: 0s - loss: 0.0038 - acc: 0.9986Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0047 - acc: 0.9985 - val_loss: 0.9414 - val_acc: 0.8240 Epoch 16/35 6475/6680 [============================>.] - ETA: 0s - loss: 0.0052 - acc: 0.9991Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 132us/step - loss: 0.0051 - acc: 0.9991 - val_loss: 0.9303 - val_acc: 0.8311 Epoch 17/35 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0036 - acc: 0.9986Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0041 - acc: 0.9985 - val_loss: 0.9874 - val_acc: 0.8299 Epoch 18/35 6253/6680 [===========================>..] - ETA: 0s - loss: 0.0038 - acc: 0.9986Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0041 - acc: 0.9985 - val_loss: 0.9847 - val_acc: 0.8347 Epoch 19/35 6290/6680 [===========================>..] - ETA: 0s - loss: 0.0040 - acc: 0.9986Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0040 - acc: 0.9985 - val_loss: 1.0185 - val_acc: 0.8323 Epoch 20/35 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0054 - acc: 0.9987Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0051 - acc: 0.9988 - val_loss: 1.0071 - val_acc: 0.8275 Epoch 21/35 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0044 - acc: 0.9987Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0045 - acc: 0.9987 - val_loss: 1.0684 - val_acc: 0.8216 Epoch 22/35 6401/6680 [===========================>..] - ETA: 0s - loss: 0.0043 - acc: 0.9983Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0041 - acc: 0.9984 - val_loss: 1.0118 - val_acc: 0.8287 Epoch 23/35 6253/6680 [===========================>..] - ETA: 0s - loss: 0.0044 - acc: 0.9987Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0043 - acc: 0.9987 - val_loss: 1.0673 - val_acc: 0.8204 Epoch 24/35 6660/6680 [============================>.] - ETA: 0s - loss: 0.0043 - acc: 0.9986Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0043 - acc: 0.9987 - val_loss: 1.0225 - val_acc: 0.8347 Epoch 25/35 6290/6680 [===========================>..] - ETA: 0s - loss: 0.0048 - acc: 0.9984Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0045 - acc: 0.9985 - val_loss: 1.0789 - val_acc: 0.8287 Epoch 26/35 6512/6680 [============================>.] - ETA: 0s - loss: 0.0051 - acc: 0.9986Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0050 - acc: 0.9987 - val_loss: 1.0682 - val_acc: 0.8323 Epoch 27/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.0045 - acc: 0.9988Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 1s 135us/step - loss: 0.0044 - acc: 0.9988 - val_loss: 1.0933 - val_acc: 0.8251 Epoch 28/35 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0049 - acc: 0.9987Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0049 - acc: 0.9987 - val_loss: 1.0988 - val_acc: 0.8299 Epoch 29/35 6290/6680 [===========================>..] - ETA: 0s - loss: 0.0035 - acc: 0.9989Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0036 - acc: 0.9988 - val_loss: 1.1008 - val_acc: 0.8299 Epoch 30/35 6660/6680 [============================>.] - ETA: 0s - loss: 0.0040 - acc: 0.9986Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 1s 135us/step - loss: 0.0040 - acc: 0.9987 - val_loss: 1.1175 - val_acc: 0.8263 Epoch 31/35 6475/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9986Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 1s 132us/step - loss: 0.0041 - acc: 0.9987 - val_loss: 1.1128 - val_acc: 0.8287 Epoch 32/35 6623/6680 [============================>.] - ETA: 0s - loss: 0.0033 - acc: 0.9988Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0042 - acc: 0.9987 - val_loss: 1.1077 - val_acc: 0.8311 Epoch 33/35 6364/6680 [===========================>..] - ETA: 0s - loss: 0.0048 - acc: 0.9986Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0046 - acc: 0.9987 - val_loss: 1.1202 - val_acc: 0.8407 Epoch 34/35 6364/6680 [===========================>..] - ETA: 0s - loss: 0.0048 - acc: 0.9986Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0046 - acc: 0.9987 - val_loss: 1.1183 - val_acc: 0.8335 Epoch 35/35 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0036 - acc: 0.9987Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0039 - acc: 0.9987 - val_loss: 1.1260 - val_acc: 0.8263 Batch size=37 Epoch=50 Train on 6680 samples, validate on 835 samples Epoch 1/50 6623/6680 [============================>.] - ETA: 0s - loss: 0.0142 - acc: 0.9973Epoch 00001: val_loss improved from inf to 0.76251, saving model to saved_models3/weights.best.ResNet_bs37_ep50.hdf5 6680/6680 [==============================] - 1s 136us/step - loss: 0.0142 - acc: 0.9973 - val_loss: 0.7625 - val_acc: 0.8192 Epoch 2/50 6290/6680 [===========================>..] - ETA: 0s - loss: 0.0095 - acc: 0.9981Epoch 00002: val_loss improved from 0.76251 to 0.71774, saving model to saved_models3/weights.best.ResNet_bs37_ep50.hdf5 6680/6680 [==============================] - 1s 135us/step - loss: 0.0103 - acc: 0.9979 - val_loss: 0.7177 - val_acc: 0.8168 Epoch 3/50 6364/6680 [===========================>..] - ETA: 0s - loss: 0.0078 - acc: 0.9978Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0076 - acc: 0.9979 - val_loss: 0.7522 - val_acc: 0.8335 Epoch 4/50 6401/6680 [===========================>..] - ETA: 0s - loss: 0.0080 - acc: 0.9981Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0078 - acc: 0.9982 - val_loss: 0.7696 - val_acc: 0.8204 Epoch 5/50 6364/6680 [===========================>..] - ETA: 0s - loss: 0.0061 - acc: 0.9987Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0075 - acc: 0.9985 - val_loss: 0.7974 - val_acc: 0.8311 Epoch 6/50 6290/6680 [===========================>..] - ETA: 0s - loss: 0.0052 - acc: 0.9989Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0052 - acc: 0.9988 - val_loss: 0.8255 - val_acc: 0.8084 Epoch 7/50 6623/6680 [============================>.] - ETA: 0s - loss: 0.0049 - acc: 0.9991Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0050 - acc: 0.9991 - val_loss: 0.8289 - val_acc: 0.8359 Epoch 8/50 6623/6680 [============================>.] - ETA: 0s - loss: 0.0052 - acc: 0.9982Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0051 - acc: 0.9982 - val_loss: 0.8849 - val_acc: 0.8347 Epoch 9/50 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0056 - acc: 0.9991Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0055 - acc: 0.9990 - val_loss: 0.8558 - val_acc: 0.8251 Epoch 10/50 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0049 - acc: 0.9986Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0047 - acc: 0.9987 - val_loss: 0.8951 - val_acc: 0.8323 Epoch 11/50 6438/6680 [===========================>..] - ETA: 0s - loss: 0.0034 - acc: 0.9984Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0040 - acc: 0.9984 - val_loss: 0.8832 - val_acc: 0.8192 Epoch 12/50 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0046 - acc: 0.9987Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0045 - acc: 0.9988 - val_loss: 0.8951 - val_acc: 0.8371 Epoch 13/50 6438/6680 [===========================>..] - ETA: 0s - loss: 0.0037 - acc: 0.9984Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0036 - acc: 0.9985 - val_loss: 0.9310 - val_acc: 0.8168 Epoch 14/50 6253/6680 [===========================>..] - ETA: 0s - loss: 0.0039 - acc: 0.9986Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0036 - acc: 0.9987 - val_loss: 0.9645 - val_acc: 0.8216 Epoch 15/50 6290/6680 [===========================>..] - ETA: 0s - loss: 0.0043 - acc: 0.9989Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0040 - acc: 0.9990 - val_loss: 0.9709 - val_acc: 0.8299 Epoch 16/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.0035 - acc: 0.9986Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0039 - acc: 0.9985 - val_loss: 1.0027 - val_acc: 0.8240 Epoch 17/50 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0031 - acc: 0.9991Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0034 - acc: 0.9990 - val_loss: 1.0029 - val_acc: 0.8335 Epoch 18/50 6364/6680 [===========================>..] - ETA: 0s - loss: 0.0050 - acc: 0.9986Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0048 - acc: 0.9987 - val_loss: 1.0110 - val_acc: 0.8299 Epoch 19/50 6290/6680 [===========================>..] - ETA: 0s - loss: 0.0031 - acc: 0.9989Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0036 - acc: 0.9987 - val_loss: 0.9973 - val_acc: 0.8287 Epoch 20/50 6623/6680 [============================>.] - ETA: 0s - loss: 0.0044 - acc: 0.9986Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 135us/step - loss: 0.0043 - acc: 0.9987 - val_loss: 1.0710 - val_acc: 0.8204 Epoch 21/50 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0030 - acc: 0.9987Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0029 - acc: 0.9988 - val_loss: 1.0188 - val_acc: 0.8287 Epoch 22/50 6364/6680 [===========================>..] - ETA: 0s - loss: 0.0032 - acc: 0.9992Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0042 - acc: 0.9990 - val_loss: 1.0491 - val_acc: 0.8251 Epoch 23/50 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0034 - acc: 0.9984Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0036 - acc: 0.9984 - val_loss: 1.0740 - val_acc: 0.8323 Epoch 24/50 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0044 - acc: 0.9987Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0046 - acc: 0.9987 - val_loss: 1.0605 - val_acc: 0.8311 Epoch 25/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.0045 - acc: 0.9985Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0045 - acc: 0.9985 - val_loss: 1.0493 - val_acc: 0.8347 Epoch 26/50 6253/6680 [===========================>..] - ETA: 0s - loss: 0.0038 - acc: 0.9986Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0045 - acc: 0.9985 - val_loss: 1.0886 - val_acc: 0.8347 Epoch 27/50 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0044 - acc: 0.9987Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0045 - acc: 0.9987 - val_loss: 1.1156 - val_acc: 0.8407 Epoch 28/50 6364/6680 [===========================>..] - ETA: 0s - loss: 0.0034 - acc: 0.9991Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0035 - acc: 0.9990 - val_loss: 1.0845 - val_acc: 0.8371 Epoch 29/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.0044 - acc: 0.9986Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0044 - acc: 0.9987 - val_loss: 1.1058 - val_acc: 0.8335 Epoch 30/50 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0039 - acc: 0.9989Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0037 - acc: 0.9990 - val_loss: 1.1136 - val_acc: 0.8383 Epoch 31/50 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0043 - acc: 0.9989Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0042 - acc: 0.9988 - val_loss: 1.1169 - val_acc: 0.8335 Epoch 32/50 6290/6680 [===========================>..] - ETA: 0s - loss: 0.0034 - acc: 0.9992Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0047 - acc: 0.9991 - val_loss: 1.1050 - val_acc: 0.8287 Epoch 33/50 6549/6680 [============================>.] - ETA: 0s - loss: 0.0048 - acc: 0.9988Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 1s 135us/step - loss: 0.0047 - acc: 0.9988 - val_loss: 1.1129 - val_acc: 0.8371 Epoch 34/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9989Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 1s 135us/step - loss: 0.0042 - acc: 0.9990 - val_loss: 1.1301 - val_acc: 0.8311 Epoch 35/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.0039 - acc: 0.9985Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0039 - acc: 0.9985 - val_loss: 1.1366 - val_acc: 0.8323 Epoch 36/50 6253/6680 [===========================>..] - ETA: 0s - loss: 0.0050 - acc: 0.9984Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0047 - acc: 0.9985 - val_loss: 1.1299 - val_acc: 0.8335 Epoch 37/50 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0043 - acc: 0.9989Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0041 - acc: 0.9990 - val_loss: 1.1458 - val_acc: 0.8323 Epoch 38/50 6364/6680 [===========================>..] - ETA: 0s - loss: 0.0043 - acc: 0.9984Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0041 - acc: 0.9985 - val_loss: 1.1422 - val_acc: 0.8251 Epoch 39/50 6290/6680 [===========================>..] - ETA: 0s - loss: 0.0022 - acc: 0.9992Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0034 - acc: 0.9990 - val_loss: 1.1293 - val_acc: 0.8419 Epoch 40/50 6401/6680 [===========================>..] - ETA: 0s - loss: 0.0030 - acc: 0.9989Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0034 - acc: 0.9988 - val_loss: 1.1550 - val_acc: 0.8407 Epoch 41/50 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0040 - acc: 0.9984Epoch 00041: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0040 - acc: 0.9984 - val_loss: 1.1714 - val_acc: 0.8347 Epoch 42/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.0040 - acc: 0.9988Epoch 00042: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0040 - acc: 0.9988 - val_loss: 1.1361 - val_acc: 0.8359 Epoch 43/50 6253/6680 [===========================>..] - ETA: 0s - loss: 0.0041 - acc: 0.9986Epoch 00043: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0039 - acc: 0.9987 - val_loss: 1.1735 - val_acc: 0.8347 Epoch 44/50 6364/6680 [===========================>..] - ETA: 0s - loss: 0.0047 - acc: 0.9987Epoch 00044: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0045 - acc: 0.9988 - val_loss: 1.1676 - val_acc: 0.8323 Epoch 45/50 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0048 - acc: 0.9986Epoch 00045: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0045 - acc: 0.9987 - val_loss: 1.1603 - val_acc: 0.8359 Epoch 46/50 6401/6680 [===========================>..] - ETA: 0s - loss: 0.0049 - acc: 0.9988Epoch 00046: val_loss did not improve 6680/6680 [==============================] - 1s 132us/step - loss: 0.0047 - acc: 0.9988 - val_loss: 1.1709 - val_acc: 0.8323 Epoch 47/50 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0046 - acc: 0.9991Epoch 00047: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0050 - acc: 0.9990 - val_loss: 1.1700 - val_acc: 0.8359 Epoch 48/50 6327/6680 [===========================>..] - ETA: 0s - loss: 0.0045 - acc: 0.9987Epoch 00048: val_loss did not improve 6680/6680 [==============================] - 1s 133us/step - loss: 0.0043 - acc: 0.9988 - val_loss: 1.2004 - val_acc: 0.8299 Epoch 49/50 6290/6680 [===========================>..] - ETA: 0s - loss: 0.0023 - acc: 0.9990Epoch 00049: val_loss did not improve 6680/6680 [==============================] - 1s 134us/step - loss: 0.0040 - acc: 0.9987 - val_loss: 1.1821 - val_acc: 0.8323 Epoch 50/50 6290/6680 [===========================>..] - ETA: 0s - loss: 0.0045 - acc: 0.9984Epoch 00050: val_loss did not improve 6680/6680 [==============================] - 1s 135us/step - loss: 0.0042 - acc: 0.9985 - val_loss: 1.1684 - val_acc: 0.8335 Batch size=40 Epoch=25 Train on 6680 samples, validate on 835 samples Epoch 1/25 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0108 - acc: 0.9976Epoch 00001: val_loss improved from inf to 0.73717, saving model to saved_models3/weights.best.ResNet_bs40_ep25.hdf5 6680/6680 [==============================] - 1s 126us/step - loss: 0.0104 - acc: 0.9978 - val_loss: 0.7372 - val_acc: 0.8311 Epoch 2/25 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0054 - acc: 0.9986Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0062 - acc: 0.9982 - val_loss: 0.8063 - val_acc: 0.8263 Epoch 3/25 6640/6680 [============================>.] - ETA: 0s - loss: 0.0060 - acc: 0.9980Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 1s 126us/step - loss: 0.0060 - acc: 0.9981 - val_loss: 0.7689 - val_acc: 0.8323 Epoch 4/25 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0058 - acc: 0.9989Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0055 - acc: 0.9990 - val_loss: 0.8289 - val_acc: 0.8383 Epoch 5/25 6640/6680 [============================>.] - ETA: 0s - loss: 0.0057 - acc: 0.9985Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 127us/step - loss: 0.0056 - acc: 0.9985 - val_loss: 0.8278 - val_acc: 0.8335 Epoch 6/25 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0048 - acc: 0.9986Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0046 - acc: 0.9987 - val_loss: 0.8454 - val_acc: 0.8299 Epoch 7/25 6640/6680 [============================>.] - ETA: 0s - loss: 0.0045 - acc: 0.9989Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0045 - acc: 0.9990 - val_loss: 0.8417 - val_acc: 0.8407 Epoch 8/25 6640/6680 [============================>.] - ETA: 0s - loss: 0.0052 - acc: 0.9985Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 126us/step - loss: 0.0051 - acc: 0.9985 - val_loss: 0.8926 - val_acc: 0.8347 Epoch 9/25 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0047 - acc: 0.9989Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0054 - acc: 0.9987 - val_loss: 0.8808 - val_acc: 0.8287 Epoch 10/25 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0041 - acc: 0.9987Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0043 - acc: 0.9987 - val_loss: 0.9091 - val_acc: 0.8347 Epoch 11/25 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0033 - acc: 0.9987Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0040 - acc: 0.9987 - val_loss: 0.9342 - val_acc: 0.8216 Epoch 12/25 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0048 - acc: 0.9984Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0044 - acc: 0.9985 - val_loss: 0.9339 - val_acc: 0.8347 Epoch 13/25 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0047 - acc: 0.9986Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0044 - acc: 0.9987 - val_loss: 0.9797 - val_acc: 0.8240 Epoch 14/25 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0031 - acc: 0.9989Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0043 - acc: 0.9987 - val_loss: 0.9991 - val_acc: 0.8216 Epoch 15/25 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0037 - acc: 0.9986Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0035 - acc: 0.9987 - val_loss: 0.9855 - val_acc: 0.8287 Epoch 16/25 6200/6680 [==========================>...] - ETA: 0s - loss: 0.0050 - acc: 0.9984Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0047 - acc: 0.9985 - val_loss: 0.9909 - val_acc: 0.8287 Epoch 17/25 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0056 - acc: 0.9987Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0056 - acc: 0.9987 - val_loss: 0.9862 - val_acc: 0.8347 Epoch 18/25 6640/6680 [============================>.] - ETA: 0s - loss: 0.0040 - acc: 0.9983Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 126us/step - loss: 0.0040 - acc: 0.9984 - val_loss: 1.0087 - val_acc: 0.8287 Epoch 19/25 6640/6680 [============================>.] - ETA: 0s - loss: 0.0045 - acc: 0.9985Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0044 - acc: 0.9985 - val_loss: 0.9995 - val_acc: 0.8311 Epoch 20/25 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0045 - acc: 0.9984Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0045 - acc: 0.9984 - val_loss: 1.0517 - val_acc: 0.8263 Epoch 21/25 6360/6680 [===========================>..] - ETA: 0s - loss: 0.0050 - acc: 0.9989Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0048 - acc: 0.9990 - val_loss: 1.0567 - val_acc: 0.8311 Epoch 22/25 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0043 - acc: 0.9987Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0044 - acc: 0.9987 - val_loss: 1.0583 - val_acc: 0.8311 Epoch 23/25 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0044 - acc: 0.9990Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0049 - acc: 0.9990 - val_loss: 1.0795 - val_acc: 0.8287 Epoch 24/25 6640/6680 [============================>.] - ETA: 0s - loss: 0.0041 - acc: 0.9985Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0041 - acc: 0.9985 - val_loss: 1.0794 - val_acc: 0.8347 Epoch 25/25 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0055 - acc: 0.9987Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0058 - acc: 0.9987 - val_loss: 1.0780 - val_acc: 0.8323 Batch size=40 Epoch=35 Train on 6680 samples, validate on 835 samples Epoch 1/35 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0076 - acc: 0.9979Epoch 00001: val_loss improved from inf to 0.80388, saving model to saved_models3/weights.best.ResNet_bs40_ep35.hdf5 6680/6680 [==============================] - 1s 127us/step - loss: 0.0086 - acc: 0.9975 - val_loss: 0.8039 - val_acc: 0.8180 Epoch 2/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.0049 - acc: 0.9988Epoch 00002: val_loss improved from 0.80388 to 0.79296, saving model to saved_models3/weights.best.ResNet_bs40_ep35.hdf5 6680/6680 [==============================] - 1s 127us/step - loss: 0.0049 - acc: 0.9988 - val_loss: 0.7930 - val_acc: 0.8335 Epoch 3/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.0056 - acc: 0.9986Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 1s 126us/step - loss: 0.0056 - acc: 0.9987 - val_loss: 0.7946 - val_acc: 0.8275 Epoch 4/35 6320/6680 [===========================>..] - ETA: 0s - loss: 0.0054 - acc: 0.9983Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0058 - acc: 0.9982 - val_loss: 0.8023 - val_acc: 0.8251 Epoch 5/35 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0049 - acc: 0.9986Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0047 - acc: 0.9987 - val_loss: 0.8043 - val_acc: 0.8359 Epoch 6/35 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0039 - acc: 0.9987Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0044 - acc: 0.9987 - val_loss: 0.8354 - val_acc: 0.8335 Epoch 7/35 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0054 - acc: 0.9987Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0051 - acc: 0.9988 - val_loss: 0.9094 - val_acc: 0.8263 Epoch 8/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.0037 - acc: 0.9985Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 127us/step - loss: 0.0037 - acc: 0.9985 - val_loss: 0.9048 - val_acc: 0.8251 Epoch 9/35 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0041 - acc: 0.9990Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0046 - acc: 0.9990 - val_loss: 0.8868 - val_acc: 0.8299 Epoch 10/35 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0038 - acc: 0.9986Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0038 - acc: 0.9985 - val_loss: 0.9228 - val_acc: 0.8287 Epoch 11/35 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0040 - acc: 0.9986Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0042 - acc: 0.9985 - val_loss: 0.9553 - val_acc: 0.8275 Epoch 12/35 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0045 - acc: 0.9989Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0042 - acc: 0.9990 - val_loss: 0.9490 - val_acc: 0.8323 Epoch 13/35 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0053 - acc: 0.9987Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0050 - acc: 0.9988 - val_loss: 0.9520 - val_acc: 0.8335 Epoch 14/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.0035 - acc: 0.9986Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 126us/step - loss: 0.0039 - acc: 0.9985 - val_loss: 0.9968 - val_acc: 0.8240 Epoch 15/35 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0032 - acc: 0.9987Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0047 - acc: 0.9987 - val_loss: 0.9726 - val_acc: 0.8311 Epoch 16/35 6320/6680 [===========================>..] - ETA: 0s - loss: 0.0042 - acc: 0.9987Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0040 - acc: 0.9988 - val_loss: 0.9934 - val_acc: 0.8335 Epoch 17/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.0043 - acc: 0.9988Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 126us/step - loss: 0.0042 - acc: 0.9988 - val_loss: 1.0123 - val_acc: 0.8251 Epoch 18/35 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0047 - acc: 0.9984Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 126us/step - loss: 0.0045 - acc: 0.9985 - val_loss: 1.0188 - val_acc: 0.8371 Epoch 19/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.0038 - acc: 0.9988Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 126us/step - loss: 0.0038 - acc: 0.9988 - val_loss: 1.0137 - val_acc: 0.8299 Epoch 20/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9988Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 126us/step - loss: 0.0042 - acc: 0.9988 - val_loss: 1.0264 - val_acc: 0.8335 Epoch 21/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.0045 - acc: 0.9986Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 1s 126us/step - loss: 0.0044 - acc: 0.9987 - val_loss: 1.0637 - val_acc: 0.8359 Epoch 22/35 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0049 - acc: 0.9986Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0046 - acc: 0.9987 - val_loss: 1.0752 - val_acc: 0.8216 Epoch 23/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.0036 - acc: 0.9986Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0036 - acc: 0.9987 - val_loss: 1.0810 - val_acc: 0.8216 Epoch 24/35 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0026 - acc: 0.9987Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0036 - acc: 0.9982 - val_loss: 1.1024 - val_acc: 0.8263 Epoch 25/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9986Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 1s 126us/step - loss: 0.0046 - acc: 0.9985 - val_loss: 1.1003 - val_acc: 0.8287 Epoch 26/35 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0041 - acc: 0.9986Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0038 - acc: 0.9987 - val_loss: 1.0903 - val_acc: 0.8359 Epoch 27/35 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0049 - acc: 0.9987Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0046 - acc: 0.9988 - val_loss: 1.0987 - val_acc: 0.8299 Epoch 28/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.0041 - acc: 0.9988Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0041 - acc: 0.9988 - val_loss: 1.1139 - val_acc: 0.8192 Epoch 29/35 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0038 - acc: 0.9986Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0042 - acc: 0.9985 - val_loss: 1.1074 - val_acc: 0.8275 Epoch 30/35 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0035 - acc: 0.9989Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0038 - acc: 0.9988 - val_loss: 1.0980 - val_acc: 0.8383 Epoch 31/35 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0044 - acc: 0.9989Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0049 - acc: 0.9988 - val_loss: 1.0906 - val_acc: 0.8323 Epoch 32/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.0044 - acc: 0.9988Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0044 - acc: 0.9988 - val_loss: 1.1301 - val_acc: 0.8335 Epoch 33/35 6320/6680 [===========================>..] - ETA: 0s - loss: 0.0045 - acc: 0.9987Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0044 - acc: 0.9987 - val_loss: 1.1357 - val_acc: 0.8299 Epoch 34/35 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0038 - acc: 0.9987Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0042 - acc: 0.9985 - val_loss: 1.1444 - val_acc: 0.8335 Epoch 35/35 6480/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9986Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 1s 131us/step - loss: 0.0041 - acc: 0.9987 - val_loss: 1.1301 - val_acc: 0.8347 Batch size=40 Epoch=50 Train on 6680 samples, validate on 835 samples Epoch 1/50 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0058 - acc: 0.9986Epoch 00001: val_loss improved from inf to 0.81031, saving model to saved_models3/weights.best.ResNet_bs40_ep50.hdf5 6680/6680 [==============================] - 1s 126us/step - loss: 0.0065 - acc: 0.9985 - val_loss: 0.8103 - val_acc: 0.8347 Epoch 2/50 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0047 - acc: 0.9989Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0058 - acc: 0.9987 - val_loss: 0.8251 - val_acc: 0.8263 Epoch 3/50 6400/6680 [===========================>..] - ETA: 0s - loss: 0.0049 - acc: 0.9980Epoch 00003: val_loss improved from 0.81031 to 0.79688, saving model to saved_models3/weights.best.ResNet_bs40_ep50.hdf5 6680/6680 [==============================] - 1s 126us/step - loss: 0.0047 - acc: 0.9981 - val_loss: 0.7969 - val_acc: 0.8311 Epoch 4/50 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0038 - acc: 0.9987Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0048 - acc: 0.9987 - val_loss: 0.8555 - val_acc: 0.8323 Epoch 5/50 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0039 - acc: 0.9987Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0046 - acc: 0.9985 - val_loss: 0.8406 - val_acc: 0.8287 Epoch 6/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.0038 - acc: 0.9986Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 126us/step - loss: 0.0040 - acc: 0.9985 - val_loss: 0.8918 - val_acc: 0.8204 Epoch 7/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9985Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0042 - acc: 0.9985 - val_loss: 0.9074 - val_acc: 0.8287 Epoch 8/50 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0057 - acc: 0.9989Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0057 - acc: 0.9988 - val_loss: 0.9107 - val_acc: 0.8347 Epoch 9/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.0037 - acc: 0.9988Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0036 - acc: 0.9988 - val_loss: 0.9490 - val_acc: 0.8275 Epoch 10/50 6320/6680 [===========================>..] - ETA: 0s - loss: 0.0044 - acc: 0.9984Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0042 - acc: 0.9985 - val_loss: 0.9256 - val_acc: 0.8299 Epoch 11/50 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0043 - acc: 0.9986Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0042 - acc: 0.9985 - val_loss: 0.9633 - val_acc: 0.8251 Epoch 12/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.0043 - acc: 0.9986Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0042 - acc: 0.9987 - val_loss: 0.9927 - val_acc: 0.8359 Epoch 13/50 6320/6680 [===========================>..] - ETA: 0s - loss: 0.0046 - acc: 0.9987Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0043 - acc: 0.9988 - val_loss: 0.9725 - val_acc: 0.8311 Epoch 14/50 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0037 - acc: 0.9990Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0041 - acc: 0.9990 - val_loss: 1.0220 - val_acc: 0.8251 Epoch 15/50 6320/6680 [===========================>..] - ETA: 0s - loss: 0.0045 - acc: 0.9981Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0042 - acc: 0.9982 - val_loss: 0.9953 - val_acc: 0.8263 Epoch 16/50 6320/6680 [===========================>..] - ETA: 0s - loss: 0.0049 - acc: 0.9987Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0047 - acc: 0.9988 - val_loss: 1.0549 - val_acc: 0.8347 Epoch 17/50 6320/6680 [===========================>..] - ETA: 0s - loss: 0.0048 - acc: 0.9987Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0046 - acc: 0.9988 - val_loss: 1.0134 - val_acc: 0.8359 Epoch 18/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.0033 - acc: 0.9988Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0036 - acc: 0.9987 - val_loss: 1.0655 - val_acc: 0.8359 Epoch 19/50 6320/6680 [===========================>..] - ETA: 0s - loss: 0.0033 - acc: 0.9992Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0035 - acc: 0.9990 - val_loss: 1.1034 - val_acc: 0.8359 Epoch 20/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.0039 - acc: 0.9985Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0039 - acc: 0.9985 - val_loss: 1.0709 - val_acc: 0.8359 Epoch 21/50 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0030 - acc: 0.9989Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0046 - acc: 0.9985 - val_loss: 1.0838 - val_acc: 0.8335 Epoch 22/50 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0044 - acc: 0.9987Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0044 - acc: 0.9987 - val_loss: 1.0821 - val_acc: 0.8251 Epoch 23/50 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0052 - acc: 0.9989Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0048 - acc: 0.9990 - val_loss: 1.0890 - val_acc: 0.8347 Epoch 24/50 6360/6680 [===========================>..] - ETA: 0s - loss: 0.0046 - acc: 0.9984Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0044 - acc: 0.9985 - val_loss: 1.0852 - val_acc: 0.8275 Epoch 25/50 6320/6680 [===========================>..] - ETA: 0s - loss: 0.0034 - acc: 0.9986Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0036 - acc: 0.9985 - val_loss: 1.1274 - val_acc: 0.8359 Epoch 26/50 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0039 - acc: 0.9984Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0036 - acc: 0.9985 - val_loss: 1.1086 - val_acc: 0.8263 Epoch 27/50 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0042 - acc: 0.9984Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0039 - acc: 0.9985 - val_loss: 1.0962 - val_acc: 0.8323 Epoch 28/50 6360/6680 [===========================>..] - ETA: 0s - loss: 0.0043 - acc: 0.9986Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 1s 123us/step - loss: 0.0041 - acc: 0.9987 - val_loss: 1.0974 - val_acc: 0.8311 Epoch 29/50 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0047 - acc: 0.9987Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0044 - acc: 0.9988 - val_loss: 1.0957 - val_acc: 0.8359 Epoch 30/50 6320/6680 [===========================>..] - ETA: 0s - loss: 0.0032 - acc: 0.9987Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0037 - acc: 0.9985 - val_loss: 1.1428 - val_acc: 0.8311 Epoch 31/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.0047 - acc: 0.9989Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0047 - acc: 0.9990 - val_loss: 1.1139 - val_acc: 0.8287 Epoch 32/50 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0048 - acc: 0.9987Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0045 - acc: 0.9988 - val_loss: 1.1197 - val_acc: 0.8347 Epoch 33/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.0045 - acc: 0.9986Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0044 - acc: 0.9987 - val_loss: 1.1323 - val_acc: 0.8359 Epoch 34/50 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0049 - acc: 0.9986Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0046 - acc: 0.9987 - val_loss: 1.1308 - val_acc: 0.8311 Epoch 35/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.0039 - acc: 0.9988Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0039 - acc: 0.9988 - val_loss: 1.1554 - val_acc: 0.8383 Epoch 36/50 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0046 - acc: 0.9984Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0043 - acc: 0.9985 - val_loss: 1.1325 - val_acc: 0.8347 Epoch 37/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.0048 - acc: 0.9988Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0048 - acc: 0.9988 - val_loss: 1.1392 - val_acc: 0.8359 Epoch 38/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.0051 - acc: 0.9988Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0051 - acc: 0.9988 - val_loss: 1.1422 - val_acc: 0.8335 Epoch 39/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.0048 - acc: 0.9986Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0047 - acc: 0.9987 - val_loss: 1.1545 - val_acc: 0.8335 Epoch 40/50 6360/6680 [===========================>..] - ETA: 0s - loss: 0.0046 - acc: 0.9989Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 1s 123us/step - loss: 0.0044 - acc: 0.9990 - val_loss: 1.1448 - val_acc: 0.8299 Epoch 41/50 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0041 - acc: 0.9989Epoch 00041: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0044 - acc: 0.9988 - val_loss: 1.1606 - val_acc: 0.8299 Epoch 42/50 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0055 - acc: 0.9987Epoch 00042: val_loss did not improve 6680/6680 [==============================] - 1s 123us/step - loss: 0.0051 - acc: 0.9988 - val_loss: 1.1482 - val_acc: 0.8347 Epoch 43/50 6320/6680 [===========================>..] - ETA: 0s - loss: 0.0031 - acc: 0.9989Epoch 00043: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0035 - acc: 0.9988 - val_loss: 1.1467 - val_acc: 0.8275 Epoch 44/50 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0042 - acc: 0.9990Epoch 00044: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0047 - acc: 0.9990 - val_loss: 1.1490 - val_acc: 0.8347 Epoch 45/50 6200/6680 [==========================>...] - ETA: 0s - loss: 0.0054 - acc: 0.9987Epoch 00045: val_loss did not improve 6680/6680 [==============================] - 1s 125us/step - loss: 0.0051 - acc: 0.9988 - val_loss: 1.1517 - val_acc: 0.8287 Epoch 46/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.0044 - acc: 0.9986Epoch 00046: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0044 - acc: 0.9987 - val_loss: 1.1644 - val_acc: 0.8323 Epoch 47/50 6280/6680 [===========================>..] - ETA: 0s - loss: 0.0047 - acc: 0.9987Epoch 00047: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0044 - acc: 0.9988 - val_loss: 1.1522 - val_acc: 0.8359 Epoch 48/50 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0047 - acc: 0.9989Epoch 00048: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0048 - acc: 0.9988 - val_loss: 1.1585 - val_acc: 0.8359 Epoch 49/50 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0041 - acc: 0.9992Epoch 00049: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0048 - acc: 0.9990 - val_loss: 1.1489 - val_acc: 0.8383 Epoch 50/50 6240/6680 [===========================>..] - ETA: 0s - loss: 0.0050 - acc: 0.9987Epoch 00050: val_loss did not improve 6680/6680 [==============================] - 1s 124us/step - loss: 0.0047 - acc: 0.9988 - val_loss: 1.1496 - val_acc: 0.8335 Batch size=64 Epoch=25 Train on 6680 samples, validate on 835 samples Epoch 1/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0050 - acc: 0.9983Epoch 00001: val_loss improved from inf to 0.88103, saving model to saved_models3/weights.best.ResNet_bs64_ep25.hdf5 6680/6680 [==============================] - 1s 89us/step - loss: 0.0048 - acc: 0.9984 - val_loss: 0.8810 - val_acc: 0.8287 Epoch 2/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0047 - acc: 0.9989Epoch 00002: val_loss improved from 0.88103 to 0.86433, saving model to saved_models3/weights.best.ResNet_bs64_ep25.hdf5 6680/6680 [==============================] - 1s 89us/step - loss: 0.0046 - acc: 0.9990 - val_loss: 0.8643 - val_acc: 0.8287 Epoch 3/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0039 - acc: 0.9989Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0040 - acc: 0.9988 - val_loss: 0.8650 - val_acc: 0.8359 Epoch 4/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9986Epoch 00004: val_loss improved from 0.86433 to 0.86284, saving model to saved_models3/weights.best.ResNet_bs64_ep25.hdf5 6680/6680 [==============================] - 1s 89us/step - loss: 0.0041 - acc: 0.9987 - val_loss: 0.8628 - val_acc: 0.8299 Epoch 5/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0040 - acc: 0.9986Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0039 - acc: 0.9987 - val_loss: 0.9286 - val_acc: 0.8228 Epoch 6/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0051 - acc: 0.9983Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0050 - acc: 0.9984 - val_loss: 0.8897 - val_acc: 0.8251 Epoch 7/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0025 - acc: 0.9991Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0040 - acc: 0.9988 - val_loss: 0.9256 - val_acc: 0.8359 Epoch 8/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0025 - acc: 0.9989Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0037 - acc: 0.9987 - val_loss: 0.9814 - val_acc: 0.8323 Epoch 9/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0031 - acc: 0.9988Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0039 - acc: 0.9987 - val_loss: 0.9467 - val_acc: 0.8323 Epoch 10/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0043 - acc: 0.9988Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0042 - acc: 0.9988 - val_loss: 0.9696 - val_acc: 0.8263 Epoch 11/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0045 - acc: 0.9986Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0043 - acc: 0.9987 - val_loss: 0.9480 - val_acc: 0.8347 Epoch 12/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0043 - acc: 0.9986Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0042 - acc: 0.9987 - val_loss: 0.9868 - val_acc: 0.8323 Epoch 13/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0037 - acc: 0.9988Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0036 - acc: 0.9988 - val_loss: 1.0207 - val_acc: 0.8323 Epoch 14/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0043 - acc: 0.9985Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0041 - acc: 0.9985 - val_loss: 0.9974 - val_acc: 0.8275 Epoch 15/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0045 - acc: 0.9985Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0044 - acc: 0.9985 - val_loss: 1.0057 - val_acc: 0.8275 Epoch 16/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9989Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0042 - acc: 0.9988 - val_loss: 1.0640 - val_acc: 0.8287 Epoch 17/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0036 - acc: 0.9986Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0041 - acc: 0.9985 - val_loss: 1.0650 - val_acc: 0.8263 Epoch 18/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0034 - acc: 0.9989Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0042 - acc: 0.9988 - val_loss: 1.0656 - val_acc: 0.8168 Epoch 19/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0036 - acc: 0.9988Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0035 - acc: 0.9988 - val_loss: 1.0248 - val_acc: 0.8335 Epoch 20/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9986Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0041 - acc: 0.9987 - val_loss: 1.0464 - val_acc: 0.8347 Epoch 21/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9986Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0041 - acc: 0.9987 - val_loss: 1.0585 - val_acc: 0.8347 Epoch 22/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0038 - acc: 0.9989Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0042 - acc: 0.9988 - val_loss: 1.0583 - val_acc: 0.8311 Epoch 23/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0037 - acc: 0.9988 Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0039 - acc: 0.9987 - val_loss: 1.1181 - val_acc: 0.8251 Epoch 24/25 6464/6680 [============================>.] - ETA: 0s - loss: 0.0032 - acc: 0.9991Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0034 - acc: 0.9990 - val_loss: 1.0509 - val_acc: 0.8299 Epoch 25/25 6528/6680 [============================>.] - ETA: 0s - loss: 0.0045 - acc: 0.9983Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0044 - acc: 0.9984 - val_loss: 1.0717 - val_acc: 0.8347 Batch size=64 Epoch=35 Train on 6680 samples, validate on 835 samples Epoch 1/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0044 - acc: 0.9988 Epoch 00001: val_loss improved from inf to 0.87991, saving model to saved_models3/weights.best.ResNet_bs64_ep35.hdf5 6680/6680 [==============================] - 1s 89us/step - loss: 0.0043 - acc: 0.9988 - val_loss: 0.8799 - val_acc: 0.8383 Epoch 2/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0048 - acc: 0.9986Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0046 - acc: 0.9987 - val_loss: 0.8959 - val_acc: 0.8347 Epoch 3/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0038 - acc: 0.9983Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0037 - acc: 0.9984 - val_loss: 0.8818 - val_acc: 0.8347 Epoch 4/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0043 - acc: 0.9988Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0041 - acc: 0.9988 - val_loss: 0.9509 - val_acc: 0.8287 Epoch 5/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0034 - acc: 0.9989Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0039 - acc: 0.9988 - val_loss: 0.9423 - val_acc: 0.8263 Epoch 6/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0046 - acc: 0.9986Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0044 - acc: 0.9987 - val_loss: 0.9511 - val_acc: 0.8359 Epoch 7/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0048 - acc: 0.9988Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0046 - acc: 0.9988 - val_loss: 0.9902 - val_acc: 0.8263 Epoch 8/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0046 - acc: 0.9986Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0045 - acc: 0.9987 - val_loss: 0.9793 - val_acc: 0.8263 Epoch 9/35 6592/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9986Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0042 - acc: 0.9987 - val_loss: 1.0159 - val_acc: 0.8275 Epoch 10/35 6528/6680 [============================>.] - ETA: 0s - loss: 0.0046 - acc: 0.9989Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0045 - acc: 0.9990 - val_loss: 1.0142 - val_acc: 0.8335 Epoch 11/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0043 - acc: 0.9988 Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0041 - acc: 0.9988 - val_loss: 1.0106 - val_acc: 0.8371 Epoch 12/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0040 - acc: 0.9988Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0039 - acc: 0.9988 - val_loss: 1.0243 - val_acc: 0.8311 Epoch 13/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0035 - acc: 0.9988Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0041 - acc: 0.9985 - val_loss: 1.0250 - val_acc: 0.8323 Epoch 14/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9985Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0042 - acc: 0.9984 - val_loss: 1.0267 - val_acc: 0.8371 Epoch 15/35 6528/6680 [============================>.] - ETA: 0s - loss: 0.0041 - acc: 0.9986Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0040 - acc: 0.9987 - val_loss: 1.0757 - val_acc: 0.8359 Epoch 16/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0040 - acc: 0.9989Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0039 - acc: 0.9990 - val_loss: 1.0430 - val_acc: 0.8371 Epoch 17/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0039 - acc: 0.9989Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0046 - acc: 0.9988 - val_loss: 1.0543 - val_acc: 0.8383 Epoch 18/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0038 - acc: 0.9989Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0037 - acc: 0.9990 - val_loss: 1.0554 - val_acc: 0.8323 Epoch 19/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0037 - acc: 0.9986Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0036 - acc: 0.9987 - val_loss: 1.0523 - val_acc: 0.8371 Epoch 20/35 6528/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9988Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0045 - acc: 0.9987 - val_loss: 1.0741 - val_acc: 0.8263 Epoch 21/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0039 - acc: 0.9988Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 1s 86us/step - loss: 0.0038 - acc: 0.9988 - val_loss: 1.0691 - val_acc: 0.8395 Epoch 22/35 6528/6680 [============================>.] - ETA: 0s - loss: 0.0041 - acc: 0.9986Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0041 - acc: 0.9987 - val_loss: 1.0950 - val_acc: 0.8323 Epoch 23/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0044 - acc: 0.9986Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0042 - acc: 0.9987 - val_loss: 1.0761 - val_acc: 0.8311 Epoch 24/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0043 - acc: 0.9989 Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0042 - acc: 0.9990 - val_loss: 1.0821 - val_acc: 0.8347 Epoch 25/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0046 - acc: 0.9991 Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0045 - acc: 0.9991 - val_loss: 1.0882 - val_acc: 0.8311 Epoch 26/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0039 - acc: 0.9991Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0043 - acc: 0.9990 - val_loss: 1.0878 - val_acc: 0.8371 Epoch 27/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0044 - acc: 0.9985Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0043 - acc: 0.9985 - val_loss: 1.1159 - val_acc: 0.8263 Epoch 28/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0040 - acc: 0.9986Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0038 - acc: 0.9987 - val_loss: 1.1045 - val_acc: 0.8383 Epoch 29/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0046 - acc: 0.9988Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0045 - acc: 0.9988 - val_loss: 1.0874 - val_acc: 0.8383 Epoch 30/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9988Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0041 - acc: 0.9988 - val_loss: 1.0989 - val_acc: 0.8395 Epoch 31/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0043 - acc: 0.9983Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0042 - acc: 0.9984 - val_loss: 1.1087 - val_acc: 0.8383 Epoch 32/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0043 - acc: 0.9989Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0041 - acc: 0.9990 - val_loss: 1.1008 - val_acc: 0.8311 Epoch 33/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0046 - acc: 0.9986Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0044 - acc: 0.9987 - val_loss: 1.0954 - val_acc: 0.8359 Epoch 34/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9986Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0046 - acc: 0.9985 - val_loss: 1.0991 - val_acc: 0.8419 Epoch 35/35 6464/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9989Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0040 - acc: 0.9990 - val_loss: 1.1156 - val_acc: 0.8347 Batch size=64 Epoch=50 Train on 6680 samples, validate on 835 samples Epoch 1/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0043 - acc: 0.9988Epoch 00001: val_loss improved from inf to 0.93283, saving model to saved_models3/weights.best.ResNet_bs64_ep50.hdf5 6680/6680 [==============================] - 1s 90us/step - loss: 0.0041 - acc: 0.9988 - val_loss: 0.9328 - val_acc: 0.8311 Epoch 2/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0040 - acc: 0.9989Epoch 00002: val_loss improved from 0.93283 to 0.91360, saving model to saved_models3/weights.best.ResNet_bs64_ep50.hdf5 6680/6680 [==============================] - 1s 91us/step - loss: 0.0039 - acc: 0.9990 - val_loss: 0.9136 - val_acc: 0.8311 Epoch 3/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0052 - acc: 0.9988Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0050 - acc: 0.9988 - val_loss: 0.9582 - val_acc: 0.8263 Epoch 4/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9988 Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 1s 89us/step - loss: 0.0041 - acc: 0.9988 - val_loss: 0.9269 - val_acc: 0.8335 Epoch 5/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0041 - acc: 0.9989Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0044 - acc: 0.9988 - val_loss: 0.9460 - val_acc: 0.8323 Epoch 6/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0043 - acc: 0.9989Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0041 - acc: 0.9990 - val_loss: 1.0013 - val_acc: 0.8299 Epoch 7/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0036 - acc: 0.9985Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 1s 89us/step - loss: 0.0035 - acc: 0.9985 - val_loss: 0.9837 - val_acc: 0.8359 Epoch 8/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0044 - acc: 0.9986Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0042 - acc: 0.9987 - val_loss: 0.9974 - val_acc: 0.8240 Epoch 9/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0050 - acc: 0.9985Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0048 - acc: 0.9985 - val_loss: 0.9789 - val_acc: 0.8335 Epoch 10/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0051 - acc: 0.9991Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0049 - acc: 0.9991 - val_loss: 1.0252 - val_acc: 0.8371 Epoch 11/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0036 - acc: 0.9986 Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0035 - acc: 0.9987 - val_loss: 1.0105 - val_acc: 0.8275 Epoch 12/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0031 - acc: 0.9988Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0035 - acc: 0.9987 - val_loss: 1.0039 - val_acc: 0.8347 Epoch 13/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0029 - acc: 0.9991Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0044 - acc: 0.9988 - val_loss: 1.0660 - val_acc: 0.8323 Epoch 14/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0025 - acc: 0.9992 Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0038 - acc: 0.9990 - val_loss: 1.0445 - val_acc: 0.8323 Epoch 15/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0036 - acc: 0.9988Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0039 - acc: 0.9987 - val_loss: 1.0358 - val_acc: 0.8287 Epoch 16/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0044 - acc: 0.9988Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0043 - acc: 0.9988 - val_loss: 1.0603 - val_acc: 0.8347 Epoch 17/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0043 - acc: 0.9986Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0042 - acc: 0.9987 - val_loss: 1.0503 - val_acc: 0.8359 Epoch 18/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0049 - acc: 0.9988Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 1s 89us/step - loss: 0.0048 - acc: 0.9988 - val_loss: 1.0584 - val_acc: 0.8431 Epoch 19/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0034 - acc: 0.9986 Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 1s 89us/step - loss: 0.0040 - acc: 0.9985 - val_loss: 1.0955 - val_acc: 0.8383 Epoch 20/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0038 - acc: 0.9989 Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0042 - acc: 0.9988 - val_loss: 1.0851 - val_acc: 0.8347 Epoch 21/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0043 - acc: 0.9989Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0041 - acc: 0.9990 - val_loss: 1.0992 - val_acc: 0.8311 Epoch 22/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9986Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0041 - acc: 0.9987 - val_loss: 1.0696 - val_acc: 0.8347 Epoch 23/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0041 - acc: 0.9986Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0040 - acc: 0.9987 - val_loss: 1.0945 - val_acc: 0.8311 Epoch 24/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0037 - acc: 0.9985Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0036 - acc: 0.9985 - val_loss: 1.0882 - val_acc: 0.8299 Epoch 25/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9989Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 1s 89us/step - loss: 0.0047 - acc: 0.9988 - val_loss: 1.0977 - val_acc: 0.8371 Epoch 26/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0041 - acc: 0.9985Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0045 - acc: 0.9984 - val_loss: 1.0964 - val_acc: 0.8335 Epoch 27/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0047 - acc: 0.9989Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0045 - acc: 0.9990 - val_loss: 1.1041 - val_acc: 0.8383 Epoch 28/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0045 - acc: 0.9989Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 1s 89us/step - loss: 0.0044 - acc: 0.9990 - val_loss: 1.0819 - val_acc: 0.8347 Epoch 29/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0049 - acc: 0.9985 Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0048 - acc: 0.9985 - val_loss: 1.0925 - val_acc: 0.8347 Epoch 30/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0046 - acc: 0.9983Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 1s 89us/step - loss: 0.0044 - acc: 0.9984 - val_loss: 1.1030 - val_acc: 0.8323 Epoch 31/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0041 - acc: 0.9988Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0040 - acc: 0.9988 - val_loss: 1.1165 - val_acc: 0.8407 Epoch 32/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0044 - acc: 0.9986Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0043 - acc: 0.9987 - val_loss: 1.1095 - val_acc: 0.8371 Epoch 33/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0033 - acc: 0.9988Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0042 - acc: 0.9987 - val_loss: 1.1279 - val_acc: 0.8335 Epoch 34/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0047 - acc: 0.9986Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 1s 89us/step - loss: 0.0046 - acc: 0.9987 - val_loss: 1.1017 - val_acc: 0.8383 Epoch 35/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0043 - acc: 0.9985Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0042 - acc: 0.9985 - val_loss: 1.1132 - val_acc: 0.8383 Epoch 36/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0044 - acc: 0.9986Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0043 - acc: 0.9987 - val_loss: 1.1072 - val_acc: 0.8347 Epoch 37/50 6528/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9988Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0047 - acc: 0.9987 - val_loss: 1.1212 - val_acc: 0.8371 Epoch 38/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0047 - acc: 0.9986Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0045 - acc: 0.9987 - val_loss: 1.1192 - val_acc: 0.8371 Epoch 39/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0039 - acc: 0.9988Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0041 - acc: 0.9987 - val_loss: 1.1143 - val_acc: 0.8371 Epoch 40/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0041 - acc: 0.9989Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0040 - acc: 0.9990 - val_loss: 1.0998 - val_acc: 0.8395 Epoch 41/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0045 - acc: 0.9988Epoch 00041: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0043 - acc: 0.9988 - val_loss: 1.1230 - val_acc: 0.8323 Epoch 42/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0047 - acc: 0.9989Epoch 00042: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0045 - acc: 0.9990 - val_loss: 1.1361 - val_acc: 0.8383 Epoch 43/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0033 - acc: 0.9991Epoch 00043: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0039 - acc: 0.9990 - val_loss: 1.1207 - val_acc: 0.8419 Epoch 44/50 6528/6680 [============================>.] - ETA: 0s - loss: 0.0042 - acc: 0.9988Epoch 00044: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0041 - acc: 0.9988 - val_loss: 1.1273 - val_acc: 0.8311 Epoch 45/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0046 - acc: 0.9988Epoch 00045: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0049 - acc: 0.9987 - val_loss: 1.1272 - val_acc: 0.8299 Epoch 46/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0039 - acc: 0.9991Epoch 00046: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0038 - acc: 0.9991 - val_loss: 1.1509 - val_acc: 0.8275 Epoch 47/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0032 - acc: 0.9991Epoch 00047: val_loss did not improve 6680/6680 [==============================] - 1s 88us/step - loss: 0.0043 - acc: 0.9990 - val_loss: 1.1262 - val_acc: 0.8395 Epoch 48/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0040 - acc: 0.9991Epoch 00048: val_loss did not improve 6680/6680 [==============================] - 1s 89us/step - loss: 0.0039 - acc: 0.9991 - val_loss: 1.1215 - val_acc: 0.8383 Epoch 49/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0040 - acc: 0.9991Epoch 00049: val_loss did not improve 6680/6680 [==============================] - 1s 89us/step - loss: 0.0044 - acc: 0.9990 - val_loss: 1.1473 - val_acc: 0.8299 Epoch 50/50 6464/6680 [============================>.] - ETA: 0s - loss: 0.0039 - acc: 0.9988Epoch 00050: val_loss did not improve 6680/6680 [==============================] - 1s 87us/step - loss: 0.0040 - acc: 0.9987 - val_loss: 1.1230 - val_acc: 0.8323
In [34]:
pd.DataFrame(fitingdict_ResNet)
Out[34]:
| Batch_Size | Epochs | Test_Accuracy | |
|---|---|---|---|
| 0 | 35 | 25 | 80.263158 |
| 1 | 35 | 35 | 82.296651 |
| 2 | 35 | 50 | 81.937799 |
| 3 | 37 | 25 | 81.937799 |
| 4 | 37 | 35 | 82.057416 |
| 5 | 37 | 50 | 80.861244 |
| 6 | 40 | 25 | 82.416268 |
| 7 | 40 | 35 | 81.818182 |
| 8 | 40 | 50 | 81.937799 |
| 9 | 64 | 25 | 82.177033 |
| 10 | 64 | 35 | 81.339713 |
| 11 | 64 | 50 | 81.937799 |
In [35]:
#take largest testaccuracy's batch size and epochs
ind=fitingdict_ResNet['Test_Accuracy'].index(max(fitingdict_ResNet['Test_Accuracy']))
bs=fitingdict_ResNet['Batch_Size'][ind]
ep=fitingdict_ResNet['Epochs'][ind]
#LOAD the model with Best validation loss
ResNet_model.load_weights('saved_models3/weights.best.ResNet_bs'+str(bs)+'_ep'+str(ep)+'.hdf5')
### TODO: Calculate classification accuracy on the test dataset.
# get index of predicted dog breed for each image in test set
ResNet_predictions = [np.argmax(ResNet_model.predict(np.expand_dims(feature, axis=0))) for feature in test_Resnet50]
# report test accuracy
test_accuracy = 100*np.sum(np.array(ResNet_predictions)==np.argmax(test_targets, axis=1))/len(ResNet_predictions)
print('Test accuracy: %.4f%%' % test_accuracy)
Test accuracy: 82.4163%
lets do with Xception model¶
In [26]:
### TODO: Obtain bottleneck features from another pre-trained CNN.
bottleneck_features = np.load('/data/bottleneck_features/DogXceptionData.npz')
train_Xception = bottleneck_features['train']
valid_Xception = bottleneck_features['valid']
test_Xception = bottleneck_features['test']
(IMPLEMENTATION) Model Architecture¶
Create a CNN to classify dog breed. At the end of your code cell block, summarize the layers of your model by executing the line:
<your model's name>.summary()
Question 5: Outline the steps you took to get to your final CNN architecture and your reasoning at each step. Describe why you think the architecture is suitable for the current problem.
Answer: The below architecture using Xception model is most suitable for current problem as it gave me Test accuracy of 86.8421% which is higher then other model i tested. We can see that in ResNet-50 the Maximum Test accuracy is 82.4163% on batch_size=40 and epoch=25 and VGG-16 is 53.1100% on batch_size=35 and epoch=40 and in Xception model the Minimum test accuracy is starts from 84.330144%. thats the reason i go with Xception model. I used with BatchNormalization as Training Deep Neural Networks is complicated by the fact that the distribution of each layer's inputs changes during training, as the parameters of the previous layers change. This slows down the training by requiring lower learning rates and careful parameter initialization, and makes it notoriously hard to train models with saturating nonlinearities. We refer to this phenomenon as internal covariate shift, and address the problem by normalizing layer inputs.
In [27]:
### TODO: Define your architecture.
from keras.layers import Conv2D, MaxPooling2D, GlobalAveragePooling2D,BatchNormalization
from keras.layers import Dropout, Flatten, Dense, Activation
from keras.models import Sequential
from keras.callbacks import ModelCheckpoint
Xception_model=Sequential()
Xception_model.add(GlobalAveragePooling2D(input_shape=train_Xception.shape[1:]))
Xception_model.add(BatchNormalization())
# let's add a fully-connected layer
Xception_model.add(Dropout(0.5))
Xception_model.add( Dense(1024, activation='relu'))
Xception_model.add( Dropout(0.5))
# and a logistic layer -- let's say we have NUM_CLASSES classes
Xception_model.add( Dense(len(dog_names), activation='softmax'))
Xception_model.summary()
_________________________________________________________________ Layer (type) Output Shape Param # ================================================================= global_average_pooling2d_3 ( (None, 2048) 0 _________________________________________________________________ batch_normalization_1 (Batch (None, 2048) 8192 _________________________________________________________________ dropout_1 (Dropout) (None, 2048) 0 _________________________________________________________________ dense_3 (Dense) (None, 1024) 2098176 _________________________________________________________________ dropout_2 (Dropout) (None, 1024) 0 _________________________________________________________________ dense_4 (Dense) (None, 133) 136325 ================================================================= Total params: 2,242,693 Trainable params: 2,238,597 Non-trainable params: 4,096 _________________________________________________________________
(IMPLEMENTATION) Compile the Model¶
In [28]:
### TODO: Compile the model
Xception_model.compile(loss="categorical_crossentropy",optimizer="rmsprop",metrics=["accuracy"])
(IMPLEMENTATION) Train the Model¶
Train your model in the code cell below. Use model checkpointing to save the model that attains the best validation loss.
You are welcome to augment the training data, but this is not a requirement.
In [55]:
### TODO: Train the model.
from sklearn.model_selection import GridSearchCV
# defining the grid search parameters
batch_size = [35,36,37,40,41]
epochs = [35,37,40,50,55]
fitingdict={'Batch_Size':[],
'Epochs':[],
'Test_Accuracy':[]}
for bs in batch_size:
for ep in epochs:
checkpointer = ModelCheckpoint(filepath='saved_models2/weights.best.ResNet_bs'+str(bs)+'_ep'+str(ep)+'.hdf5',
verbose=1, save_best_only=True)
print("\nBatch size={0} Epoch={1}".format(bs,ep))
Xception_model.fit(train_Xception, train_targets,validation_data=(valid_Xception, valid_targets),
epochs=ep , batch_size=bs,
callbacks=[checkpointer],verbose=1)
#LOAD the model with Best validation loss
Xception_model.load_weights('saved_models2/weights.best.ResNet_bs'+str(bs)+'_ep'+str(ep)+'.hdf5')
Xception_predictions = [np.argmax(Xception_model.predict(np.expand_dims(feature, axis=0))) for feature in test_Xception]
test_accuracy = 100*np.sum(np.array(Xception_predictions)==np.argmax(test_targets, axis=1))/len(Xception_predictions)
fitingdict['Batch_Size'].append(bs)
fitingdict['Epochs'].append(ep)
fitingdict['Test_Accuracy'].append(test_accuracy)
Batch size=35 Epoch=35 Train on 6680 samples, validate on 835 samples Epoch 1/35 6615/6680 [============================>.] - ETA: 0s - loss: 0.1983 - acc: 0.9708Epoch 00001: val_loss improved from inf to 1.46264, saving model to saved_models2/weights.best.ResNet_bs35_ep35.hdf5 6680/6680 [==============================] - 4s 558us/step - loss: 0.1990 - acc: 0.9707 - val_loss: 1.4626 - val_acc: 0.8587 Epoch 2/35 6615/6680 [============================>.] - ETA: 0s - loss: 0.2076 - acc: 0.9701Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.2091 - acc: 0.9699 - val_loss: 1.5414 - val_acc: 0.8479 Epoch 3/35 6650/6680 [============================>.] - ETA: 0s - loss: 0.2040 - acc: 0.9677Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 4s 565us/step - loss: 0.2032 - acc: 0.9678 - val_loss: 1.5532 - val_acc: 0.8455 Epoch 4/35 6580/6680 [============================>.] - ETA: 0s - loss: 0.1886 - acc: 0.9702Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 4s 548us/step - loss: 0.1905 - acc: 0.9702 - val_loss: 1.5364 - val_acc: 0.8599 Epoch 5/35 6650/6680 [============================>.] - ETA: 0s - loss: 0.2214 - acc: 0.9686Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.2214 - acc: 0.9686 - val_loss: 1.5839 - val_acc: 0.8431 Epoch 6/35 6615/6680 [============================>.] - ETA: 0s - loss: 0.2250 - acc: 0.9675Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.2228 - acc: 0.9678 - val_loss: 1.5875 - val_acc: 0.8491 Epoch 7/35 6580/6680 [============================>.] - ETA: 0s - loss: 0.2215 - acc: 0.9687Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 4s 544us/step - loss: 0.2210 - acc: 0.9687 - val_loss: 1.5775 - val_acc: 0.8479 Epoch 8/35 6615/6680 [============================>.] - ETA: 0s - loss: 0.2353 - acc: 0.9672Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 4s 548us/step - loss: 0.2337 - acc: 0.9674 - val_loss: 1.5295 - val_acc: 0.8635 Epoch 9/35 6580/6680 [============================>.] - ETA: 0s - loss: 0.2365 - acc: 0.9649Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 4s 547us/step - loss: 0.2396 - acc: 0.9650 - val_loss: 1.5912 - val_acc: 0.8563 Epoch 10/35 6615/6680 [============================>.] - ETA: 0s - loss: 0.2046 - acc: 0.9708Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 4s 542us/step - loss: 0.2043 - acc: 0.9708 - val_loss: 1.6751 - val_acc: 0.8491 Epoch 11/35 6615/6680 [============================>.] - ETA: 0s - loss: 0.2118 - acc: 0.9681Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 4s 546us/step - loss: 0.2098 - acc: 0.9684 - val_loss: 1.6789 - val_acc: 0.8539 Epoch 12/35 6650/6680 [============================>.] - ETA: 0s - loss: 0.2341 - acc: 0.9669Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.2338 - acc: 0.9668 - val_loss: 1.6250 - val_acc: 0.8539 Epoch 13/35 6650/6680 [============================>.] - ETA: 0s - loss: 0.2284 - acc: 0.9693Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 4s 547us/step - loss: 0.2277 - acc: 0.9693 - val_loss: 1.6745 - val_acc: 0.8383 Epoch 14/35 6580/6680 [============================>.] - ETA: 0s - loss: 0.2067 - acc: 0.9691Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 4s 553us/step - loss: 0.2037 - acc: 0.9696 - val_loss: 1.6129 - val_acc: 0.8539 Epoch 15/35 6580/6680 [============================>.] - ETA: 0s - loss: 0.2221 - acc: 0.9691Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.2221 - acc: 0.9693 - val_loss: 1.6107 - val_acc: 0.8563 Epoch 16/35 6580/6680 [============================>.] - ETA: 0s - loss: 0.2156 - acc: 0.9705Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.2158 - acc: 0.9705 - val_loss: 1.7286 - val_acc: 0.8395 Epoch 17/35 6650/6680 [============================>.] - ETA: 0s - loss: 0.2373 - acc: 0.9686Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.2387 - acc: 0.9686 - val_loss: 1.6144 - val_acc: 0.8539 Epoch 18/35 6615/6680 [============================>.] - ETA: 0s - loss: 0.2206 - acc: 0.9687Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 4s 545us/step - loss: 0.2193 - acc: 0.9686 - val_loss: 1.5646 - val_acc: 0.8515 Epoch 19/35 6580/6680 [============================>.] - ETA: 0s - loss: 0.2287 - acc: 0.9673Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 4s 542us/step - loss: 0.2288 - acc: 0.9674 - val_loss: 1.5570 - val_acc: 0.8515 Epoch 20/35 6580/6680 [============================>.] - ETA: 0s - loss: 0.2382 - acc: 0.9667Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 4s 554us/step - loss: 0.2354 - acc: 0.9671 - val_loss: 1.6206 - val_acc: 0.8491 Epoch 21/35 6580/6680 [============================>.] - ETA: 0s - loss: 0.2268 - acc: 0.9682Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 4s 547us/step - loss: 0.2269 - acc: 0.9680 - val_loss: 1.5614 - val_acc: 0.8539 Epoch 22/35 6650/6680 [============================>.] - ETA: 0s - loss: 0.2028 - acc: 0.9698Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 4s 546us/step - loss: 0.2034 - acc: 0.9698 - val_loss: 1.6625 - val_acc: 0.8527 Epoch 23/35 6615/6680 [============================>.] - ETA: 0s - loss: 0.2289 - acc: 0.9690Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 4s 548us/step - loss: 0.2269 - acc: 0.9692 - val_loss: 1.6965 - val_acc: 0.8503 Epoch 24/35 6615/6680 [============================>.] - ETA: 0s - loss: 0.2574 - acc: 0.9643Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 4s 550us/step - loss: 0.2549 - acc: 0.9647 - val_loss: 1.7081 - val_acc: 0.8491 Epoch 25/35 6580/6680 [============================>.] - ETA: 0s - loss: 0.2560 - acc: 0.9684Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 4s 545us/step - loss: 0.2553 - acc: 0.9683 - val_loss: 1.7107 - val_acc: 0.8551 Epoch 26/35 6580/6680 [============================>.] - ETA: 0s - loss: 0.2156 - acc: 0.9690Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 4s 544us/step - loss: 0.2155 - acc: 0.9692 - val_loss: 1.7609 - val_acc: 0.8539 Epoch 27/35 6650/6680 [============================>.] - ETA: 0s - loss: 0.2194 - acc: 0.9698Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 4s 544us/step - loss: 0.2208 - acc: 0.9698 - val_loss: 1.7678 - val_acc: 0.8407 Epoch 28/35 6650/6680 [============================>.] - ETA: 0s - loss: 0.2147 - acc: 0.9708Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.2162 - acc: 0.9707 - val_loss: 1.7648 - val_acc: 0.8431 Epoch 29/35 6650/6680 [============================>.] - ETA: 0s - loss: 0.2105 - acc: 0.9723Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.2096 - acc: 0.9725 - val_loss: 1.7228 - val_acc: 0.8455 Epoch 30/35 6650/6680 [============================>.] - ETA: 0s - loss: 0.2411 - acc: 0.9695Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.2424 - acc: 0.9695 - val_loss: 1.6352 - val_acc: 0.8491 Epoch 31/35 6650/6680 [============================>.] - ETA: 0s - loss: 0.2312 - acc: 0.9689Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 4s 547us/step - loss: 0.2317 - acc: 0.9689 - val_loss: 1.6527 - val_acc: 0.8443 Epoch 32/35 6615/6680 [============================>.] - ETA: 0s - loss: 0.2396 - acc: 0.9684Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 4s 554us/step - loss: 0.2386 - acc: 0.9686 - val_loss: 1.7082 - val_acc: 0.8479 Epoch 33/35 6650/6680 [============================>.] - ETA: 0s - loss: 0.2295 - acc: 0.9704Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 4s 553us/step - loss: 0.2303 - acc: 0.9702 - val_loss: 1.6698 - val_acc: 0.8587 Epoch 34/35 6650/6680 [============================>.] - ETA: 0s - loss: 0.2157 - acc: 0.9722Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 4s 560us/step - loss: 0.2148 - acc: 0.9723 - val_loss: 1.6363 - val_acc: 0.8551 Epoch 35/35 6650/6680 [============================>.] - ETA: 0s - loss: 0.1835 - acc: 0.9747Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 4s 545us/step - loss: 0.1839 - acc: 0.9746 - val_loss: 1.7155 - val_acc: 0.8479 Batch size=35 Epoch=37 Train on 6680 samples, validate on 835 samples Epoch 1/37 6580/6680 [============================>.] - ETA: 0s - loss: 0.2034 - acc: 0.9688Epoch 00001: val_loss improved from inf to 1.49988, saving model to saved_models2/weights.best.ResNet_bs35_ep37.hdf5 6680/6680 [==============================] - 4s 549us/step - loss: 0.2048 - acc: 0.9686 - val_loss: 1.4999 - val_acc: 0.8467 Epoch 2/37 6615/6680 [============================>.] - ETA: 0s - loss: 0.2080 - acc: 0.9692Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 4s 542us/step - loss: 0.2071 - acc: 0.9690 - val_loss: 1.5362 - val_acc: 0.8491 Epoch 3/37 6580/6680 [============================>.] - ETA: 0s - loss: 0.1951 - acc: 0.9704Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.1967 - acc: 0.9702 - val_loss: 1.6042 - val_acc: 0.8431 Epoch 4/37 6615/6680 [============================>.] - ETA: 0s - loss: 0.2176 - acc: 0.9684Epoch 00004: val_loss improved from 1.49988 to 1.47475, saving model to saved_models2/weights.best.ResNet_bs35_ep37.hdf5 6680/6680 [==============================] - 4s 533us/step - loss: 0.2204 - acc: 0.9680 - val_loss: 1.4748 - val_acc: 0.8479 Epoch 5/37 6650/6680 [============================>.] - ETA: 0s - loss: 0.2302 - acc: 0.9657Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.2330 - acc: 0.9656 - val_loss: 1.5861 - val_acc: 0.8371 Epoch 6/37 6615/6680 [============================>.] - ETA: 0s - loss: 0.2440 - acc: 0.9661Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.2422 - acc: 0.9663 - val_loss: 1.5851 - val_acc: 0.8443 Epoch 7/37 6580/6680 [============================>.] - ETA: 0s - loss: 0.2183 - acc: 0.9705Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 4s 544us/step - loss: 0.2192 - acc: 0.9704 - val_loss: 1.6422 - val_acc: 0.8431 Epoch 8/37 6580/6680 [============================>.] - ETA: 0s - loss: 0.2278 - acc: 0.9663Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 4s 549us/step - loss: 0.2276 - acc: 0.9660 - val_loss: 1.6035 - val_acc: 0.8431 Epoch 9/37 6580/6680 [============================>.] - ETA: 0s - loss: 0.2083 - acc: 0.9690Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 4s 546us/step - loss: 0.2114 - acc: 0.9689 - val_loss: 1.5379 - val_acc: 0.8479 Epoch 10/37 6650/6680 [============================>.] - ETA: 0s - loss: 0.2075 - acc: 0.9690Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 4s 546us/step - loss: 0.2076 - acc: 0.9689 - val_loss: 1.6530 - val_acc: 0.8419 Epoch 11/37 6650/6680 [============================>.] - ETA: 0s - loss: 0.2309 - acc: 0.9642Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 4s 549us/step - loss: 0.2299 - acc: 0.9644 - val_loss: 1.6663 - val_acc: 0.8539 Epoch 12/37 6615/6680 [============================>.] - ETA: 0s - loss: 0.2173 - acc: 0.9704Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 4s 551us/step - loss: 0.2177 - acc: 0.9705 - val_loss: 1.5863 - val_acc: 0.8479 Epoch 13/37 6580/6680 [============================>.] - ETA: 0s - loss: 0.2146 - acc: 0.9690Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 4s 547us/step - loss: 0.2201 - acc: 0.9684 - val_loss: 1.6451 - val_acc: 0.8467 Epoch 14/37 6650/6680 [============================>.] - ETA: 0s - loss: 0.2316 - acc: 0.9678Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 4s 550us/step - loss: 0.2326 - acc: 0.9678 - val_loss: 1.6160 - val_acc: 0.8515 Epoch 15/37 6615/6680 [============================>.] - ETA: 0s - loss: 0.2086 - acc: 0.9717Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.2070 - acc: 0.9719 - val_loss: 1.6887 - val_acc: 0.8443 Epoch 16/37 6615/6680 [============================>.] - ETA: 0s - loss: 0.2209 - acc: 0.9690Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.2228 - acc: 0.9684 - val_loss: 1.6448 - val_acc: 0.8503 Epoch 17/37 6580/6680 [============================>.] - ETA: 0s - loss: 0.2307 - acc: 0.9681Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 4s 547us/step - loss: 0.2314 - acc: 0.9680 - val_loss: 1.6715 - val_acc: 0.8479 Epoch 18/37 6650/6680 [============================>.] - ETA: 0s - loss: 0.2274 - acc: 0.9681Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 4s 543us/step - loss: 0.2272 - acc: 0.9680 - val_loss: 1.6892 - val_acc: 0.8539 Epoch 19/37 6650/6680 [============================>.] - ETA: 0s - loss: 0.2228 - acc: 0.9681Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.2223 - acc: 0.9681 - val_loss: 1.6204 - val_acc: 0.8491 Epoch 20/37 6650/6680 [============================>.] - ETA: 0s - loss: 0.2435 - acc: 0.9660Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 4s 542us/step - loss: 0.2427 - acc: 0.9660 - val_loss: 1.7267 - val_acc: 0.8527 Epoch 21/37 6580/6680 [============================>.] - ETA: 0s - loss: 0.2355 - acc: 0.9679Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 4s 551us/step - loss: 0.2349 - acc: 0.9678 - val_loss: 1.7047 - val_acc: 0.8455 Epoch 22/37 6615/6680 [============================>.] - ETA: 0s - loss: 0.1936 - acc: 0.9717Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 4s 546us/step - loss: 0.1942 - acc: 0.9719 - val_loss: 1.6707 - val_acc: 0.8479 Epoch 23/37 6615/6680 [============================>.] - ETA: 0s - loss: 0.2089 - acc: 0.9702Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.2103 - acc: 0.9701 - val_loss: 1.6694 - val_acc: 0.8491 Epoch 24/37 6580/6680 [============================>.] - ETA: 0s - loss: 0.2125 - acc: 0.9701Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 4s 541us/step - loss: 0.2124 - acc: 0.9701 - val_loss: 1.6934 - val_acc: 0.8395 Epoch 25/37 6615/6680 [============================>.] - ETA: 0s - loss: 0.2319 - acc: 0.9693Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.2345 - acc: 0.9693 - val_loss: 1.6656 - val_acc: 0.8503 Epoch 26/37 6580/6680 [============================>.] - ETA: 0s - loss: 0.1993 - acc: 0.9736Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.1976 - acc: 0.9737 - val_loss: 1.7276 - val_acc: 0.8515 Epoch 27/37 6615/6680 [============================>.] - ETA: 0s - loss: 0.2080 - acc: 0.9701Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.2090 - acc: 0.9698 - val_loss: 1.7145 - val_acc: 0.8467 Epoch 28/37 6615/6680 [============================>.] - ETA: 0s - loss: 0.2293 - acc: 0.9692Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 4s 544us/step - loss: 0.2281 - acc: 0.9693 - val_loss: 1.8158 - val_acc: 0.8491 Epoch 29/37 6615/6680 [============================>.] - ETA: 0s - loss: 0.2177 - acc: 0.9689Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 4s 544us/step - loss: 0.2191 - acc: 0.9689 - val_loss: 1.7366 - val_acc: 0.8491 Epoch 30/37 6545/6680 [============================>.] - ETA: 0s - loss: 0.2335 - acc: 0.9716Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.2313 - acc: 0.9720 - val_loss: 1.7351 - val_acc: 0.8467 Epoch 31/37 6615/6680 [============================>.] - ETA: 0s - loss: 0.1734 - acc: 0.9755Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.1741 - acc: 0.9756 - val_loss: 1.7981 - val_acc: 0.8479 Epoch 32/37 6580/6680 [============================>.] - ETA: 0s - loss: 0.2093 - acc: 0.9710Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 4s 547us/step - loss: 0.2100 - acc: 0.9711 - val_loss: 1.6974 - val_acc: 0.8503 Epoch 33/37 6650/6680 [============================>.] - ETA: 0s - loss: 0.2102 - acc: 0.9723Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 4s 554us/step - loss: 0.2109 - acc: 0.9722 - val_loss: 1.7001 - val_acc: 0.8479 Epoch 34/37 6650/6680 [============================>.] - ETA: 0s - loss: 0.1960 - acc: 0.9752Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.1951 - acc: 0.9753 - val_loss: 1.7649 - val_acc: 0.8455 Epoch 35/37 6580/6680 [============================>.] - ETA: 0s - loss: 0.1914 - acc: 0.9723Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 4s 542us/step - loss: 0.1891 - acc: 0.9726 - val_loss: 1.6935 - val_acc: 0.8503 Epoch 36/37 6615/6680 [============================>.] - ETA: 0s - loss: 0.2145 - acc: 0.9713Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.2172 - acc: 0.9713 - val_loss: 1.6966 - val_acc: 0.8479 Epoch 37/37 6615/6680 [============================>.] - ETA: 0s - loss: 0.2002 - acc: 0.9743Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.1998 - acc: 0.9744 - val_loss: 1.7340 - val_acc: 0.8419 Batch size=35 Epoch=40 Train on 6680 samples, validate on 835 samples Epoch 1/40 6650/6680 [============================>.] - ETA: 0s - loss: 0.2050 - acc: 0.9698Epoch 00001: val_loss improved from inf to 1.57961, saving model to saved_models2/weights.best.ResNet_bs35_ep40.hdf5 6680/6680 [==============================] - 4s 540us/step - loss: 0.2051 - acc: 0.9698 - val_loss: 1.5796 - val_acc: 0.8527 Epoch 2/40 6650/6680 [============================>.] - ETA: 0s - loss: 0.2181 - acc: 0.9662Epoch 00002: val_loss improved from 1.57961 to 1.57918, saving model to saved_models2/weights.best.ResNet_bs35_ep40.hdf5 6680/6680 [==============================] - 4s 541us/step - loss: 0.2183 - acc: 0.9662 - val_loss: 1.5792 - val_acc: 0.8515 Epoch 3/40 6580/6680 [============================>.] - ETA: 0s - loss: 0.2174 - acc: 0.9682Epoch 00003: val_loss improved from 1.57918 to 1.55264, saving model to saved_models2/weights.best.ResNet_bs35_ep40.hdf5 6680/6680 [==============================] - 4s 545us/step - loss: 0.2185 - acc: 0.9683 - val_loss: 1.5526 - val_acc: 0.8479 Epoch 4/40 6580/6680 [============================>.] - ETA: 0s - loss: 0.1973 - acc: 0.9707Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.1983 - acc: 0.9708 - val_loss: 1.5899 - val_acc: 0.8455 Epoch 5/40 6580/6680 [============================>.] - ETA: 0s - loss: 0.2116 - acc: 0.9690Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 4s 542us/step - loss: 0.2111 - acc: 0.9689 - val_loss: 1.6002 - val_acc: 0.8503 Epoch 6/40 6615/6680 [============================>.] - ETA: 0s - loss: 0.2186 - acc: 0.9690Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 4s 544us/step - loss: 0.2173 - acc: 0.9690 - val_loss: 1.7026 - val_acc: 0.8443 Epoch 7/40 6650/6680 [============================>.] - ETA: 0s - loss: 0.1951 - acc: 0.9698Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 4s 545us/step - loss: 0.1949 - acc: 0.9698 - val_loss: 1.6093 - val_acc: 0.8443 Epoch 8/40 6580/6680 [============================>.] - ETA: 0s - loss: 0.2294 - acc: 0.9661Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 4s 553us/step - loss: 0.2266 - acc: 0.9663 - val_loss: 1.6390 - val_acc: 0.8479 Epoch 9/40 6650/6680 [============================>.] - ETA: 0s - loss: 0.2156 - acc: 0.9698Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 4s 556us/step - loss: 0.2147 - acc: 0.9699 - val_loss: 1.7127 - val_acc: 0.8443 Epoch 10/40 6650/6680 [============================>.] - ETA: 0s - loss: 0.1985 - acc: 0.9711Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 4s 543us/step - loss: 0.1980 - acc: 0.9711 - val_loss: 1.6772 - val_acc: 0.8491 Epoch 11/40 6580/6680 [============================>.] - ETA: 0s - loss: 0.2006 - acc: 0.9684Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1978 - acc: 0.9687 - val_loss: 1.6577 - val_acc: 0.8587 Epoch 12/40 6650/6680 [============================>.] - ETA: 0s - loss: 0.2267 - acc: 0.9674Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.2259 - acc: 0.9674 - val_loss: 1.7310 - val_acc: 0.8503 Epoch 13/40 6650/6680 [============================>.] - ETA: 0s - loss: 0.2166 - acc: 0.9711Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 4s 547us/step - loss: 0.2167 - acc: 0.9710 - val_loss: 1.6735 - val_acc: 0.8515 Epoch 14/40 6615/6680 [============================>.] - ETA: 0s - loss: 0.2109 - acc: 0.9725Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 4s 555us/step - loss: 0.2105 - acc: 0.9726 - val_loss: 1.6766 - val_acc: 0.8527 Epoch 15/40 6615/6680 [============================>.] - ETA: 0s - loss: 0.2174 - acc: 0.9701Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 4s 542us/step - loss: 0.2181 - acc: 0.9701 - val_loss: 1.6891 - val_acc: 0.8539 Epoch 16/40 6650/6680 [============================>.] - ETA: 0s - loss: 0.2128 - acc: 0.9672Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 4s 543us/step - loss: 0.2145 - acc: 0.9671 - val_loss: 1.7557 - val_acc: 0.8491 Epoch 17/40 6650/6680 [============================>.] - ETA: 0s - loss: 0.1981 - acc: 0.9698Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.1975 - acc: 0.9698 - val_loss: 1.7375 - val_acc: 0.8539 Epoch 18/40 6650/6680 [============================>.] - ETA: 0s - loss: 0.2200 - acc: 0.9696Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.2204 - acc: 0.9695 - val_loss: 1.7566 - val_acc: 0.8515 Epoch 19/40 6615/6680 [============================>.] - ETA: 0s - loss: 0.2073 - acc: 0.9717Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.2093 - acc: 0.9714 - val_loss: 1.7231 - val_acc: 0.8551 Epoch 20/40 6580/6680 [============================>.] - ETA: 0s - loss: 0.2118 - acc: 0.9716Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.2114 - acc: 0.9714 - val_loss: 1.6856 - val_acc: 0.8479 Epoch 21/40 6580/6680 [============================>.] - ETA: 0s - loss: 0.2253 - acc: 0.9707Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.2313 - acc: 0.9698 - val_loss: 1.7512 - val_acc: 0.8419 Epoch 22/40 6650/6680 [============================>.] - ETA: 0s - loss: 0.2037 - acc: 0.9731Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 4s 542us/step - loss: 0.2044 - acc: 0.9731 - val_loss: 1.8039 - val_acc: 0.8443 Epoch 23/40 6580/6680 [============================>.] - ETA: 0s - loss: 0.2255 - acc: 0.9714Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.2263 - acc: 0.9713 - val_loss: 1.8122 - val_acc: 0.8419 Epoch 24/40 6650/6680 [============================>.] - ETA: 0s - loss: 0.2358 - acc: 0.9692Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.2358 - acc: 0.9692 - val_loss: 1.8351 - val_acc: 0.8395 Epoch 25/40 6580/6680 [============================>.] - ETA: 0s - loss: 0.2113 - acc: 0.9716Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.2148 - acc: 0.9716 - val_loss: 1.7378 - val_acc: 0.8539 Epoch 26/40 6615/6680 [============================>.] - ETA: 0s - loss: 0.2262 - acc: 0.9716Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 4s 548us/step - loss: 0.2240 - acc: 0.9719 - val_loss: 1.6965 - val_acc: 0.8491 Epoch 27/40 6650/6680 [============================>.] - ETA: 0s - loss: 0.2210 - acc: 0.9705Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 4s 541us/step - loss: 0.2214 - acc: 0.9704 - val_loss: 1.7100 - val_acc: 0.8479 Epoch 28/40 6580/6680 [============================>.] - ETA: 0s - loss: 0.2120 - acc: 0.9698Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 4s 549us/step - loss: 0.2177 - acc: 0.9695 - val_loss: 1.6952 - val_acc: 0.8551 Epoch 29/40 6580/6680 [============================>.] - ETA: 0s - loss: 0.1829 - acc: 0.9743Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 4s 551us/step - loss: 0.1807 - acc: 0.9746 - val_loss: 1.7109 - val_acc: 0.8491 Epoch 30/40 6615/6680 [============================>.] - ETA: 0s - loss: 0.1996 - acc: 0.9738Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.2052 - acc: 0.9735 - val_loss: 1.7434 - val_acc: 0.8563 Epoch 31/40 6580/6680 [============================>.] - ETA: 0s - loss: 0.1937 - acc: 0.9745Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.2014 - acc: 0.9737 - val_loss: 1.7621 - val_acc: 0.8503 Epoch 32/40 6580/6680 [============================>.] - ETA: 0s - loss: 0.2028 - acc: 0.9726Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.2003 - acc: 0.9728 - val_loss: 1.8297 - val_acc: 0.8479 Epoch 33/40 6650/6680 [============================>.] - ETA: 0s - loss: 0.2345 - acc: 0.9687Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.2337 - acc: 0.9687 - val_loss: 1.8895 - val_acc: 0.8419 Epoch 34/40 6615/6680 [============================>.] - ETA: 0s - loss: 0.2124 - acc: 0.9716Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.2118 - acc: 0.9716 - val_loss: 1.8027 - val_acc: 0.8503 Epoch 35/40 6580/6680 [============================>.] - ETA: 0s - loss: 0.2141 - acc: 0.9723Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.2115 - acc: 0.9726 - val_loss: 1.7901 - val_acc: 0.8479 Epoch 36/40 6615/6680 [============================>.] - ETA: 0s - loss: 0.2277 - acc: 0.9699Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.2255 - acc: 0.9702 - val_loss: 1.7508 - val_acc: 0.8503 Epoch 37/40 6615/6680 [============================>.] - ETA: 0s - loss: 0.2198 - acc: 0.9749Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 4s 548us/step - loss: 0.2185 - acc: 0.9750 - val_loss: 1.7222 - val_acc: 0.8539 Epoch 38/40 6580/6680 [============================>.] - ETA: 0s - loss: 0.2210 - acc: 0.9726Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 4s 545us/step - loss: 0.2211 - acc: 0.9726 - val_loss: 1.6103 - val_acc: 0.8671 Epoch 39/40 6580/6680 [============================>.] - ETA: 0s - loss: 0.2491 - acc: 0.9690Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 4s 546us/step - loss: 0.2551 - acc: 0.9687 - val_loss: 1.7978 - val_acc: 0.8455 Epoch 40/40 6615/6680 [============================>.] - ETA: 0s - loss: 0.2161 - acc: 0.9728Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 4s 545us/step - loss: 0.2177 - acc: 0.9725 - val_loss: 1.7624 - val_acc: 0.8551 Batch size=35 Epoch=50 Train on 6680 samples, validate on 835 samples Epoch 1/50 6615/6680 [============================>.] - ETA: 0s - loss: 0.2373 - acc: 0.9670Epoch 00001: val_loss improved from inf to 1.61396, saving model to saved_models2/weights.best.ResNet_bs35_ep50.hdf5 6680/6680 [==============================] - 4s 546us/step - loss: 0.2353 - acc: 0.9672 - val_loss: 1.6140 - val_acc: 0.8479 Epoch 2/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.2221 - acc: 0.9680Epoch 00002: val_loss improved from 1.61396 to 1.58716, saving model to saved_models2/weights.best.ResNet_bs35_ep50.hdf5 6680/6680 [==============================] - 4s 550us/step - loss: 0.2211 - acc: 0.9681 - val_loss: 1.5872 - val_acc: 0.8515 Epoch 3/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.2156 - acc: 0.9678Epoch 00003: val_loss improved from 1.58716 to 1.53811, saving model to saved_models2/weights.best.ResNet_bs35_ep50.hdf5 6680/6680 [==============================] - 4s 543us/step - loss: 0.2147 - acc: 0.9680 - val_loss: 1.5381 - val_acc: 0.8527 Epoch 4/50 6580/6680 [============================>.] - ETA: 0s - loss: 0.2352 - acc: 0.9682Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.2326 - acc: 0.9684 - val_loss: 1.5989 - val_acc: 0.8491 Epoch 5/50 6580/6680 [============================>.] - ETA: 0s - loss: 0.2225 - acc: 0.9687Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 4s 541us/step - loss: 0.2202 - acc: 0.9690 - val_loss: 1.6043 - val_acc: 0.8479 Epoch 6/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.2194 - acc: 0.9693Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.2203 - acc: 0.9693 - val_loss: 1.5950 - val_acc: 0.8491 Epoch 7/50 6580/6680 [============================>.] - ETA: 0s - loss: 0.2419 - acc: 0.9657Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 4s 542us/step - loss: 0.2438 - acc: 0.9651 - val_loss: 1.5617 - val_acc: 0.8515 Epoch 8/50 6580/6680 [============================>.] - ETA: 0s - loss: 0.1951 - acc: 0.9717Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 4s 541us/step - loss: 0.1970 - acc: 0.9719 - val_loss: 1.6485 - val_acc: 0.8563 Epoch 9/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.2196 - acc: 0.9698Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.2187 - acc: 0.9699 - val_loss: 1.6241 - val_acc: 0.8443 Epoch 10/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.2509 - acc: 0.9662Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 4s 546us/step - loss: 0.2498 - acc: 0.9663 - val_loss: 1.5574 - val_acc: 0.8563 Epoch 11/50 6580/6680 [============================>.] - ETA: 0s - loss: 0.1924 - acc: 0.9733Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.1924 - acc: 0.9734 - val_loss: 1.5948 - val_acc: 0.8563 Epoch 12/50 6580/6680 [============================>.] - ETA: 0s - loss: 0.2180 - acc: 0.9687Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.2177 - acc: 0.9684 - val_loss: 1.6430 - val_acc: 0.8551 Epoch 13/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.2280 - acc: 0.9681Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.2271 - acc: 0.9683 - val_loss: 1.5841 - val_acc: 0.8515 Epoch 14/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.2185 - acc: 0.9702Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.2176 - acc: 0.9704 - val_loss: 1.5892 - val_acc: 0.8623 Epoch 15/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.2331 - acc: 0.9677Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 4s 542us/step - loss: 0.2322 - acc: 0.9678 - val_loss: 1.5724 - val_acc: 0.8587 Epoch 16/50 6615/6680 [============================>.] - ETA: 0s - loss: 0.2091 - acc: 0.9717Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.2082 - acc: 0.9716 - val_loss: 1.6042 - val_acc: 0.8599 Epoch 17/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.2041 - acc: 0.9729Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 4s 557us/step - loss: 0.2040 - acc: 0.9729 - val_loss: 1.6621 - val_acc: 0.8539 Epoch 18/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.2191 - acc: 0.9698Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 4s 547us/step - loss: 0.2208 - acc: 0.9696 - val_loss: 1.6563 - val_acc: 0.8479 Epoch 19/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.1909 - acc: 0.9737Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 4s 544us/step - loss: 0.1901 - acc: 0.9738 - val_loss: 1.5439 - val_acc: 0.8599 Epoch 20/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.2361 - acc: 0.9689Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 4s 558us/step - loss: 0.2350 - acc: 0.9690 - val_loss: 1.6316 - val_acc: 0.8563 Epoch 21/50 6580/6680 [============================>.] - ETA: 0s - loss: 0.1697 - acc: 0.9748Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.1695 - acc: 0.9750 - val_loss: 1.7533 - val_acc: 0.8455 Epoch 22/50 6580/6680 [============================>.] - ETA: 0s - loss: 0.1951 - acc: 0.9739Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 4s 541us/step - loss: 0.1932 - acc: 0.9741 - val_loss: 1.7192 - val_acc: 0.8563 Epoch 23/50 6580/6680 [============================>.] - ETA: 0s - loss: 0.2083 - acc: 0.9714Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.2079 - acc: 0.9716 - val_loss: 1.7421 - val_acc: 0.8431 Epoch 24/50 6615/6680 [============================>.] - ETA: 0s - loss: 0.2171 - acc: 0.9695Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 4s 541us/step - loss: 0.2155 - acc: 0.9695 - val_loss: 1.6734 - val_acc: 0.8467 Epoch 25/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.2226 - acc: 0.9722Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 4s 552us/step - loss: 0.2216 - acc: 0.9723 - val_loss: 1.7285 - val_acc: 0.8431 Epoch 26/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.1990 - acc: 0.9726Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.1981 - acc: 0.9728 - val_loss: 1.7738 - val_acc: 0.8551 Epoch 27/50 6615/6680 [============================>.] - ETA: 0s - loss: 0.2141 - acc: 0.9720Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.2130 - acc: 0.9719 - val_loss: 1.8278 - val_acc: 0.8431 Epoch 28/50 6615/6680 [============================>.] - ETA: 0s - loss: 0.2176 - acc: 0.9711Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.2174 - acc: 0.9708 - val_loss: 1.7967 - val_acc: 0.8503 Epoch 29/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.2201 - acc: 0.9710Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.2202 - acc: 0.9710 - val_loss: 1.6576 - val_acc: 0.8587 Epoch 30/50 6580/6680 [============================>.] - ETA: 0s - loss: 0.2104 - acc: 0.9717Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.2148 - acc: 0.9713 - val_loss: 1.7897 - val_acc: 0.8455 Epoch 31/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.2067 - acc: 0.9722Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.2090 - acc: 0.9719 - val_loss: 1.7685 - val_acc: 0.8455 Epoch 32/50 6615/6680 [============================>.] - ETA: 0s - loss: 0.2238 - acc: 0.9695Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.2267 - acc: 0.9693 - val_loss: 1.6748 - val_acc: 0.8515 Epoch 33/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.2188 - acc: 0.9711Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.2180 - acc: 0.9711 - val_loss: 1.7607 - val_acc: 0.8443 Epoch 34/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.2052 - acc: 0.9729Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.2044 - acc: 0.9729 - val_loss: 1.8621 - val_acc: 0.8431 Epoch 35/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.2085 - acc: 0.9711Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 4s 547us/step - loss: 0.2089 - acc: 0.9710 - val_loss: 1.6788 - val_acc: 0.8551 Epoch 36/50 6545/6680 [============================>.] - ETA: 0s - loss: 0.2065 - acc: 0.9739Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.2092 - acc: 0.9735 - val_loss: 1.7225 - val_acc: 0.8419 Epoch 37/50 6580/6680 [============================>.] - ETA: 0s - loss: 0.1997 - acc: 0.9746Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.1998 - acc: 0.9746 - val_loss: 1.6924 - val_acc: 0.8575 Epoch 38/50 6615/6680 [============================>.] - ETA: 0s - loss: 0.2274 - acc: 0.9723Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 4s 545us/step - loss: 0.2255 - acc: 0.9725 - val_loss: 1.7085 - val_acc: 0.8527 Epoch 39/50 6580/6680 [============================>.] - ETA: 0s - loss: 0.1970 - acc: 0.9719Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 4s 544us/step - loss: 0.1947 - acc: 0.9722 - val_loss: 1.6784 - val_acc: 0.8587 Epoch 40/50 6615/6680 [============================>.] - ETA: 0s - loss: 0.1997 - acc: 0.9737Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.1986 - acc: 0.9738 - val_loss: 1.6745 - val_acc: 0.8575 Epoch 41/50 6615/6680 [============================>.] - ETA: 0s - loss: 0.1916 - acc: 0.9749Epoch 00041: val_loss did not improve 6680/6680 [==============================] - 4s 541us/step - loss: 0.1900 - acc: 0.9750 - val_loss: 1.6702 - val_acc: 0.8599 Epoch 42/50 6580/6680 [============================>.] - ETA: 0s - loss: 0.1986 - acc: 0.9745Epoch 00042: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.2000 - acc: 0.9744 - val_loss: 1.7238 - val_acc: 0.8563 Epoch 43/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.1906 - acc: 0.9743Epoch 00043: val_loss did not improve 6680/6680 [==============================] - 4s 542us/step - loss: 0.1921 - acc: 0.9743 - val_loss: 1.5852 - val_acc: 0.8623 Epoch 44/50 6615/6680 [============================>.] - ETA: 0s - loss: 0.2217 - acc: 0.9704Epoch 00044: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.2229 - acc: 0.9704 - val_loss: 1.7437 - val_acc: 0.8575 Epoch 45/50 6615/6680 [============================>.] - ETA: 0s - loss: 0.2064 - acc: 0.9732Epoch 00045: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.2069 - acc: 0.9732 - val_loss: 1.7420 - val_acc: 0.8527 Epoch 46/50 6580/6680 [============================>.] - ETA: 0s - loss: 0.2207 - acc: 0.9736Epoch 00046: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.2174 - acc: 0.9740 - val_loss: 1.6796 - val_acc: 0.8455 Epoch 47/50 6580/6680 [============================>.] - ETA: 0s - loss: 0.2010 - acc: 0.9751Epoch 00047: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.2011 - acc: 0.9751 - val_loss: 1.7013 - val_acc: 0.8551 Epoch 48/50 6615/6680 [============================>.] - ETA: 0s - loss: 0.1765 - acc: 0.9764Epoch 00048: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.1748 - acc: 0.9766 - val_loss: 1.7327 - val_acc: 0.8491 Epoch 49/50 6650/6680 [============================>.] - ETA: 0s - loss: 0.2010 - acc: 0.9756Epoch 00049: val_loss did not improve 6680/6680 [==============================] - 4s 550us/step - loss: 0.2027 - acc: 0.9754 - val_loss: 1.7723 - val_acc: 0.8563 Epoch 50/50 6615/6680 [============================>.] - ETA: 0s - loss: 0.2294 - acc: 0.9699Epoch 00050: val_loss did not improve 6680/6680 [==============================] - 4s 547us/step - loss: 0.2282 - acc: 0.9699 - val_loss: 1.6563 - val_acc: 0.8611 Batch size=35 Epoch=55 Train on 6680 samples, validate on 835 samples Epoch 1/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.2337 - acc: 0.9672Epoch 00001: val_loss improved from inf to 1.53496, saving model to saved_models2/weights.best.ResNet_bs35_ep55.hdf5 6680/6680 [==============================] - 4s 543us/step - loss: 0.2318 - acc: 0.9671 - val_loss: 1.5350 - val_acc: 0.8527 Epoch 2/55 6615/6680 [============================>.] - ETA: 0s - loss: 0.2314 - acc: 0.9675Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.2299 - acc: 0.9675 - val_loss: 1.5563 - val_acc: 0.8551 Epoch 3/55 6615/6680 [============================>.] - ETA: 0s - loss: 0.2377 - acc: 0.9666Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.2368 - acc: 0.9668 - val_loss: 1.6695 - val_acc: 0.8599 Epoch 4/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.2163 - acc: 0.9699Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.2172 - acc: 0.9695 - val_loss: 1.6477 - val_acc: 0.8479 Epoch 5/55 6615/6680 [============================>.] - ETA: 0s - loss: 0.2156 - acc: 0.9696Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.2187 - acc: 0.9693 - val_loss: 1.6535 - val_acc: 0.8587 Epoch 6/55 6615/6680 [============================>.] - ETA: 0s - loss: 0.2113 - acc: 0.9686Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 4s 545us/step - loss: 0.2113 - acc: 0.9683 - val_loss: 1.7155 - val_acc: 0.8407 Epoch 7/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.2318 - acc: 0.9676Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.2310 - acc: 0.9678 - val_loss: 1.6509 - val_acc: 0.8491 Epoch 8/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.2347 - acc: 0.9682Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 4s 543us/step - loss: 0.2328 - acc: 0.9684 - val_loss: 1.5887 - val_acc: 0.8467 Epoch 9/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.2383 - acc: 0.9679Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.2370 - acc: 0.9681 - val_loss: 1.6649 - val_acc: 0.8515 Epoch 10/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.2182 - acc: 0.9687Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 4s 544us/step - loss: 0.2208 - acc: 0.9687 - val_loss: 1.7140 - val_acc: 0.8479 Epoch 11/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.1808 - acc: 0.9743Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 4s 546us/step - loss: 0.1809 - acc: 0.9744 - val_loss: 1.7025 - val_acc: 0.8431 Epoch 12/55 6650/6680 [============================>.] - ETA: 0s - loss: 0.1903 - acc: 0.9729Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.1902 - acc: 0.9729 - val_loss: 1.6464 - val_acc: 0.8575 Epoch 13/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.2066 - acc: 0.9714Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.2065 - acc: 0.9716 - val_loss: 1.7272 - val_acc: 0.8551 Epoch 14/55 6615/6680 [============================>.] - ETA: 0s - loss: 0.1981 - acc: 0.9726Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 4s 545us/step - loss: 0.2004 - acc: 0.9725 - val_loss: 1.6762 - val_acc: 0.8479 Epoch 15/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.2080 - acc: 0.9690Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 4s 543us/step - loss: 0.2129 - acc: 0.9687 - val_loss: 1.6263 - val_acc: 0.8587 Epoch 16/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.2490 - acc: 0.9685Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.2516 - acc: 0.9686 - val_loss: 1.6783 - val_acc: 0.8635 Epoch 17/55 6615/6680 [============================>.] - ETA: 0s - loss: 0.2528 - acc: 0.9676Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 4s 550us/step - loss: 0.2521 - acc: 0.9675 - val_loss: 1.6556 - val_acc: 0.8539 Epoch 18/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.1925 - acc: 0.9749Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.1996 - acc: 0.9743 - val_loss: 1.6879 - val_acc: 0.8527 Epoch 19/55 6650/6680 [============================>.] - ETA: 0s - loss: 0.1925 - acc: 0.9714Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.1940 - acc: 0.9714 - val_loss: 1.7877 - val_acc: 0.8443 Epoch 20/55 6615/6680 [============================>.] - ETA: 0s - loss: 0.1928 - acc: 0.9731Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 4s 541us/step - loss: 0.1910 - acc: 0.9734 - val_loss: 1.7319 - val_acc: 0.8431 Epoch 21/55 6650/6680 [============================>.] - ETA: 0s - loss: 0.2235 - acc: 0.9680Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 4s 552us/step - loss: 0.2225 - acc: 0.9681 - val_loss: 1.6924 - val_acc: 0.8563 Epoch 22/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.2057 - acc: 0.9731Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.2052 - acc: 0.9731 - val_loss: 1.7560 - val_acc: 0.8491 Epoch 23/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.2144 - acc: 0.9701Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.2150 - acc: 0.9699 - val_loss: 1.7489 - val_acc: 0.8515 Epoch 24/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.1932 - acc: 0.9740Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.1905 - acc: 0.9744 - val_loss: 1.7383 - val_acc: 0.8503 Epoch 25/55 6650/6680 [============================>.] - ETA: 0s - loss: 0.2164 - acc: 0.9711Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.2154 - acc: 0.9713 - val_loss: 1.6781 - val_acc: 0.8479 Epoch 26/55 6650/6680 [============================>.] - ETA: 0s - loss: 0.2083 - acc: 0.9738Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.2081 - acc: 0.9738 - val_loss: 1.6529 - val_acc: 0.8539 Epoch 27/55 6650/6680 [============================>.] - ETA: 0s - loss: 0.1862 - acc: 0.9723Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.1895 - acc: 0.9720 - val_loss: 1.7055 - val_acc: 0.8491 Epoch 28/55 6650/6680 [============================>.] - ETA: 0s - loss: 0.2164 - acc: 0.9702Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 4s 545us/step - loss: 0.2176 - acc: 0.9702 - val_loss: 1.7171 - val_acc: 0.8407 Epoch 29/55 6650/6680 [============================>.] - ETA: 0s - loss: 0.1941 - acc: 0.9749Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.1939 - acc: 0.9747 - val_loss: 1.7295 - val_acc: 0.8407 Epoch 30/55 6650/6680 [============================>.] - ETA: 0s - loss: 0.2381 - acc: 0.9693Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.2375 - acc: 0.9693 - val_loss: 1.7486 - val_acc: 0.8431 Epoch 31/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.2279 - acc: 0.9708Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.2279 - acc: 0.9708 - val_loss: 1.6858 - val_acc: 0.8491 Epoch 32/55 6650/6680 [============================>.] - ETA: 0s - loss: 0.2050 - acc: 0.9725Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 4s 546us/step - loss: 0.2042 - acc: 0.9726 - val_loss: 1.8060 - val_acc: 0.8407 Epoch 33/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.1888 - acc: 0.9754Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 4s 546us/step - loss: 0.1863 - acc: 0.9756 - val_loss: 1.8319 - val_acc: 0.8371 Epoch 34/55 6650/6680 [============================>.] - ETA: 0s - loss: 0.1844 - acc: 0.9752Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.1862 - acc: 0.9750 - val_loss: 1.7339 - val_acc: 0.8467 Epoch 35/55 6650/6680 [============================>.] - ETA: 0s - loss: 0.2277 - acc: 0.9722Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.2267 - acc: 0.9723 - val_loss: 1.7243 - val_acc: 0.8563 Epoch 36/55 6545/6680 [============================>.] - ETA: 0s - loss: 0.2134 - acc: 0.9719Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.2113 - acc: 0.9719 - val_loss: 1.7754 - val_acc: 0.8539 Epoch 37/55 6615/6680 [============================>.] - ETA: 0s - loss: 0.1994 - acc: 0.9749Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.1976 - acc: 0.9750 - val_loss: 1.7738 - val_acc: 0.8527 Epoch 38/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.2227 - acc: 0.9729Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 4s 541us/step - loss: 0.2229 - acc: 0.9729 - val_loss: 1.6982 - val_acc: 0.8563 Epoch 39/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.2243 - acc: 0.9704Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.2243 - acc: 0.9705 - val_loss: 1.7920 - val_acc: 0.8503 Epoch 40/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.1941 - acc: 0.9743Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.1919 - acc: 0.9746 - val_loss: 1.7830 - val_acc: 0.8479 Epoch 41/55 6615/6680 [============================>.] - ETA: 0s - loss: 0.2062 - acc: 0.9737Epoch 00041: val_loss did not improve 6680/6680 [==============================] - 4s 544us/step - loss: 0.2061 - acc: 0.9737 - val_loss: 1.8822 - val_acc: 0.8419 Epoch 42/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.2008 - acc: 0.9751Epoch 00042: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.2003 - acc: 0.9753 - val_loss: 1.8787 - val_acc: 0.8467 Epoch 43/55 6615/6680 [============================>.] - ETA: 0s - loss: 0.2279 - acc: 0.9723Epoch 00043: val_loss did not improve 6680/6680 [==============================] - 4s 556us/step - loss: 0.2309 - acc: 0.9722 - val_loss: 1.8503 - val_acc: 0.8515 Epoch 44/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.1870 - acc: 0.9751Epoch 00044: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.1843 - acc: 0.9754 - val_loss: 1.7806 - val_acc: 0.8479 Epoch 45/55 6650/6680 [============================>.] - ETA: 0s - loss: 0.1990 - acc: 0.9734Epoch 00045: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.1982 - acc: 0.9735 - val_loss: 1.7717 - val_acc: 0.8503 Epoch 46/55 6650/6680 [============================>.] - ETA: 0s - loss: 0.2177 - acc: 0.9717Epoch 00046: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.2167 - acc: 0.9719 - val_loss: 1.8158 - val_acc: 0.8491 Epoch 47/55 6650/6680 [============================>.] - ETA: 0s - loss: 0.2193 - acc: 0.9729Epoch 00047: val_loss did not improve 6680/6680 [==============================] - 4s 541us/step - loss: 0.2185 - acc: 0.9729 - val_loss: 1.8537 - val_acc: 0.8503 Epoch 48/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.2004 - acc: 0.9736Epoch 00048: val_loss did not improve 6680/6680 [==============================] - 4s 547us/step - loss: 0.2005 - acc: 0.9735 - val_loss: 1.8280 - val_acc: 0.8527 Epoch 49/55 6615/6680 [============================>.] - ETA: 0s - loss: 0.1678 - acc: 0.9767Epoch 00049: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.1671 - acc: 0.9768 - val_loss: 1.7521 - val_acc: 0.8635 Epoch 50/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.1952 - acc: 0.9739Epoch 00050: val_loss did not improve 6680/6680 [==============================] - 4s 553us/step - loss: 0.1947 - acc: 0.9738 - val_loss: 1.7072 - val_acc: 0.8539 Epoch 51/55 6615/6680 [============================>.] - ETA: 0s - loss: 0.2219 - acc: 0.9741Epoch 00051: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.2291 - acc: 0.9737 - val_loss: 1.8146 - val_acc: 0.8467 Epoch 52/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.1993 - acc: 0.9737Epoch 00052: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.1999 - acc: 0.9737 - val_loss: 1.8970 - val_acc: 0.8431 Epoch 53/55 6580/6680 [============================>.] - ETA: 0s - loss: 0.1869 - acc: 0.9748Epoch 00053: val_loss did not improve 6680/6680 [==============================] - 4s 542us/step - loss: 0.1893 - acc: 0.9743 - val_loss: 1.8256 - val_acc: 0.8467 Epoch 54/55 6650/6680 [============================>.] - ETA: 0s - loss: 0.1895 - acc: 0.9761Epoch 00054: val_loss did not improve 6680/6680 [==============================] - 4s 551us/step - loss: 0.1892 - acc: 0.9760 - val_loss: 1.8524 - val_acc: 0.8479 Epoch 55/55 6650/6680 [============================>.] - ETA: 0s - loss: 0.1889 - acc: 0.9746Epoch 00055: val_loss did not improve 6680/6680 [==============================] - 4s 550us/step - loss: 0.1881 - acc: 0.9747 - val_loss: 1.8245 - val_acc: 0.8467 Batch size=36 Epoch=35 Train on 6680 samples, validate on 835 samples Epoch 1/35 6624/6680 [============================>.] - ETA: 0s - loss: 0.2251 - acc: 0.9706Epoch 00001: val_loss improved from inf to 1.64562, saving model to saved_models2/weights.best.ResNet_bs36_ep35.hdf5 6680/6680 [==============================] - 4s 538us/step - loss: 0.2259 - acc: 0.9705 - val_loss: 1.6456 - val_acc: 0.8563 Epoch 2/35 6624/6680 [============================>.] - ETA: 0s - loss: 0.2313 - acc: 0.9671Epoch 00002: val_loss improved from 1.64562 to 1.62733, saving model to saved_models2/weights.best.ResNet_bs36_ep35.hdf5 6680/6680 [==============================] - 4s 553us/step - loss: 0.2297 - acc: 0.9672 - val_loss: 1.6273 - val_acc: 0.8527 Epoch 3/35 6660/6680 [============================>.] - ETA: 0s - loss: 0.2041 - acc: 0.9713Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.2035 - acc: 0.9714 - val_loss: 1.6556 - val_acc: 0.8527 Epoch 4/35 6660/6680 [============================>.] - ETA: 0s - loss: 0.2205 - acc: 0.9683Epoch 00004: val_loss improved from 1.62733 to 1.59287, saving model to saved_models2/weights.best.ResNet_bs36_ep35.hdf5 6680/6680 [==============================] - 4s 553us/step - loss: 0.2221 - acc: 0.9683 - val_loss: 1.5929 - val_acc: 0.8515 Epoch 5/35 6660/6680 [============================>.] - ETA: 0s - loss: 0.2127 - acc: 0.9673Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.2125 - acc: 0.9672 - val_loss: 1.6604 - val_acc: 0.8515 Epoch 6/35 6588/6680 [============================>.] - ETA: 0s - loss: 0.1778 - acc: 0.9730Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.1809 - acc: 0.9728 - val_loss: 1.6150 - val_acc: 0.8491 Epoch 7/35 6624/6680 [============================>.] - ETA: 0s - loss: 0.2187 - acc: 0.9683Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.2174 - acc: 0.9684 - val_loss: 1.6720 - val_acc: 0.8515 Epoch 8/35 6660/6680 [============================>.] - ETA: 0s - loss: 0.1820 - acc: 0.9718Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 4s 541us/step - loss: 0.1825 - acc: 0.9717 - val_loss: 1.6518 - val_acc: 0.8455 Epoch 9/35 6624/6680 [============================>.] - ETA: 0s - loss: 0.2453 - acc: 0.9691Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.2448 - acc: 0.9690 - val_loss: 1.6650 - val_acc: 0.8467 Epoch 10/35 6588/6680 [============================>.] - ETA: 0s - loss: 0.2311 - acc: 0.9689Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.2297 - acc: 0.9687 - val_loss: 1.6412 - val_acc: 0.8503 Epoch 11/35 6588/6680 [============================>.] - ETA: 0s - loss: 0.2110 - acc: 0.9699Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 4s 541us/step - loss: 0.2172 - acc: 0.9693 - val_loss: 1.6268 - val_acc: 0.8479 Epoch 12/35 6660/6680 [============================>.] - ETA: 0s - loss: 0.2024 - acc: 0.9734Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.2022 - acc: 0.9734 - val_loss: 1.6712 - val_acc: 0.8455 Epoch 13/35 6624/6680 [============================>.] - ETA: 0s - loss: 0.2068 - acc: 0.9731Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.2090 - acc: 0.9728 - val_loss: 1.7154 - val_acc: 0.8527 Epoch 14/35 6624/6680 [============================>.] - ETA: 0s - loss: 0.2143 - acc: 0.9710Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.2126 - acc: 0.9713 - val_loss: 1.6685 - val_acc: 0.8515 Epoch 15/35 6588/6680 [============================>.] - ETA: 0s - loss: 0.2199 - acc: 0.9681Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.2203 - acc: 0.9681 - val_loss: 1.6442 - val_acc: 0.8527 Epoch 16/35 6624/6680 [============================>.] - ETA: 0s - loss: 0.2012 - acc: 0.9710Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.2022 - acc: 0.9710 - val_loss: 1.6515 - val_acc: 0.8515 Epoch 17/35 6660/6680 [============================>.] - ETA: 0s - loss: 0.2096 - acc: 0.9721Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.2090 - acc: 0.9722 - val_loss: 1.6546 - val_acc: 0.8503 Epoch 18/35 6624/6680 [============================>.] - ETA: 0s - loss: 0.2136 - acc: 0.9706Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.2143 - acc: 0.9707 - val_loss: 1.6965 - val_acc: 0.8467 Epoch 19/35 6624/6680 [============================>.] - ETA: 0s - loss: 0.2161 - acc: 0.9701Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 4s 541us/step - loss: 0.2191 - acc: 0.9701 - val_loss: 1.7219 - val_acc: 0.8503 Epoch 20/35 6624/6680 [============================>.] - ETA: 0s - loss: 0.2328 - acc: 0.9716Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.2336 - acc: 0.9714 - val_loss: 1.7442 - val_acc: 0.8503 Epoch 21/35 6660/6680 [============================>.] - ETA: 0s - loss: 0.2102 - acc: 0.9700Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.2121 - acc: 0.9698 - val_loss: 1.7989 - val_acc: 0.8431 Epoch 22/35 6624/6680 [============================>.] - ETA: 0s - loss: 0.1966 - acc: 0.9736Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.1966 - acc: 0.9735 - val_loss: 1.7029 - val_acc: 0.8575 Epoch 23/35 6660/6680 [============================>.] - ETA: 0s - loss: 0.1824 - acc: 0.9740Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.1818 - acc: 0.9741 - val_loss: 1.6775 - val_acc: 0.8527 Epoch 24/35 6624/6680 [============================>.] - ETA: 0s - loss: 0.2191 - acc: 0.9715Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.2214 - acc: 0.9713 - val_loss: 1.6912 - val_acc: 0.8551 Epoch 25/35 6660/6680 [============================>.] - ETA: 0s - loss: 0.2172 - acc: 0.9719Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.2166 - acc: 0.9720 - val_loss: 1.7433 - val_acc: 0.8527 Epoch 26/35 6588/6680 [============================>.] - ETA: 0s - loss: 0.2229 - acc: 0.9706Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.2212 - acc: 0.9705 - val_loss: 1.7523 - val_acc: 0.8431 Epoch 27/35 6588/6680 [============================>.] - ETA: 0s - loss: 0.2189 - acc: 0.9704Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.2199 - acc: 0.9704 - val_loss: 1.7139 - val_acc: 0.8551 Epoch 28/35 6588/6680 [============================>.] - ETA: 0s - loss: 0.1723 - acc: 0.9765Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1706 - acc: 0.9765 - val_loss: 1.7937 - val_acc: 0.8479 Epoch 29/35 6624/6680 [============================>.] - ETA: 0s - loss: 0.1738 - acc: 0.9758Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.1723 - acc: 0.9760 - val_loss: 1.7498 - val_acc: 0.8515 Epoch 30/35 6660/6680 [============================>.] - ETA: 0s - loss: 0.2110 - acc: 0.9751Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 4s 541us/step - loss: 0.2104 - acc: 0.9751 - val_loss: 1.7236 - val_acc: 0.8515 Epoch 31/35 6624/6680 [============================>.] - ETA: 0s - loss: 0.1849 - acc: 0.9736Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.1898 - acc: 0.9731 - val_loss: 1.6944 - val_acc: 0.8527 Epoch 32/35 6588/6680 [============================>.] - ETA: 0s - loss: 0.1787 - acc: 0.9750Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 4s 544us/step - loss: 0.1833 - acc: 0.9746 - val_loss: 1.7182 - val_acc: 0.8479 Epoch 33/35 6660/6680 [============================>.] - ETA: 0s - loss: 0.1928 - acc: 0.9752Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.1939 - acc: 0.9751 - val_loss: 1.8333 - val_acc: 0.8503 Epoch 34/35 6588/6680 [============================>.] - ETA: 0s - loss: 0.2112 - acc: 0.9712Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.2098 - acc: 0.9714 - val_loss: 1.7831 - val_acc: 0.8491 Epoch 35/35 6660/6680 [============================>.] - ETA: 0s - loss: 0.2124 - acc: 0.9734Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.2118 - acc: 0.9735 - val_loss: 1.8777 - val_acc: 0.8455 Batch size=36 Epoch=37 Train on 6680 samples, validate on 835 samples Epoch 1/37 6624/6680 [============================>.] - ETA: 0s - loss: 0.2058 - acc: 0.9722Epoch 00001: val_loss improved from inf to 1.70843, saving model to saved_models2/weights.best.ResNet_bs36_ep37.hdf5 6680/6680 [==============================] - 4s 553us/step - loss: 0.2059 - acc: 0.9720 - val_loss: 1.7084 - val_acc: 0.8503 Epoch 2/37 6624/6680 [============================>.] - ETA: 0s - loss: 0.2373 - acc: 0.9698Epoch 00002: val_loss improved from 1.70843 to 1.68580, saving model to saved_models2/weights.best.ResNet_bs36_ep37.hdf5 6680/6680 [==============================] - 4s 549us/step - loss: 0.2355 - acc: 0.9701 - val_loss: 1.6858 - val_acc: 0.8455 Epoch 3/37 6624/6680 [============================>.] - ETA: 0s - loss: 0.2329 - acc: 0.9681Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.2309 - acc: 0.9684 - val_loss: 1.7367 - val_acc: 0.8467 Epoch 4/37 6624/6680 [============================>.] - ETA: 0s - loss: 0.1982 - acc: 0.9707Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 4s 545us/step - loss: 0.1972 - acc: 0.9707 - val_loss: 1.7250 - val_acc: 0.8503 Epoch 5/37 6624/6680 [============================>.] - ETA: 0s - loss: 0.2357 - acc: 0.9675Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 4s 544us/step - loss: 0.2337 - acc: 0.9678 - val_loss: 1.6995 - val_acc: 0.8503 Epoch 6/37 6624/6680 [============================>.] - ETA: 0s - loss: 0.2482 - acc: 0.9678Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 4s 542us/step - loss: 0.2485 - acc: 0.9680 - val_loss: 1.7817 - val_acc: 0.8395 Epoch 7/37 6624/6680 [============================>.] - ETA: 0s - loss: 0.1971 - acc: 0.9743Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 4s 549us/step - loss: 0.1966 - acc: 0.9744 - val_loss: 1.7229 - val_acc: 0.8515 Epoch 8/37 6660/6680 [============================>.] - ETA: 0s - loss: 0.2249 - acc: 0.9683Epoch 00008: val_loss improved from 1.68580 to 1.63297, saving model to saved_models2/weights.best.ResNet_bs36_ep37.hdf5 6680/6680 [==============================] - 4s 552us/step - loss: 0.2242 - acc: 0.9684 - val_loss: 1.6330 - val_acc: 0.8575 Epoch 9/37 6588/6680 [============================>.] - ETA: 0s - loss: 0.1799 - acc: 0.9733Epoch 00009: val_loss improved from 1.63297 to 1.61080, saving model to saved_models2/weights.best.ResNet_bs36_ep37.hdf5 6680/6680 [==============================] - 4s 549us/step - loss: 0.1874 - acc: 0.9723 - val_loss: 1.6108 - val_acc: 0.8527 Epoch 10/37 6588/6680 [============================>.] - ETA: 0s - loss: 0.2255 - acc: 0.9698Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.2272 - acc: 0.9699 - val_loss: 1.7514 - val_acc: 0.8455 Epoch 11/37 6660/6680 [============================>.] - ETA: 0s - loss: 0.2106 - acc: 0.9716Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.2100 - acc: 0.9717 - val_loss: 1.6597 - val_acc: 0.8527 Epoch 12/37 6660/6680 [============================>.] - ETA: 0s - loss: 0.1951 - acc: 0.9746Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.1946 - acc: 0.9747 - val_loss: 1.7412 - val_acc: 0.8491 Epoch 13/37 6588/6680 [============================>.] - ETA: 0s - loss: 0.2120 - acc: 0.9698Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 4s 544us/step - loss: 0.2135 - acc: 0.9696 - val_loss: 1.6672 - val_acc: 0.8503 Epoch 14/37 6588/6680 [============================>.] - ETA: 0s - loss: 0.1911 - acc: 0.9701Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 4s 544us/step - loss: 0.1902 - acc: 0.9702 - val_loss: 1.7309 - val_acc: 0.8503 Epoch 15/37 6660/6680 [============================>.] - ETA: 0s - loss: 0.2127 - acc: 0.9701Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.2127 - acc: 0.9701 - val_loss: 1.7519 - val_acc: 0.8575 Epoch 16/37 6660/6680 [============================>.] - ETA: 0s - loss: 0.2133 - acc: 0.9680Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.2129 - acc: 0.9680 - val_loss: 1.7439 - val_acc: 0.8479 Epoch 17/37 6552/6680 [============================>.] - ETA: 0s - loss: 0.2508 - acc: 0.9666Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.2488 - acc: 0.9668 - val_loss: 1.6651 - val_acc: 0.8527 Epoch 18/37 6588/6680 [============================>.] - ETA: 0s - loss: 0.2223 - acc: 0.9677Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.2192 - acc: 0.9681 - val_loss: 1.7171 - val_acc: 0.8539 Epoch 19/37 6624/6680 [============================>.] - ETA: 0s - loss: 0.2151 - acc: 0.9731Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.2134 - acc: 0.9734 - val_loss: 1.6873 - val_acc: 0.8539 Epoch 20/37 6624/6680 [============================>.] - ETA: 0s - loss: 0.1995 - acc: 0.9740Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 4s 543us/step - loss: 0.1978 - acc: 0.9743 - val_loss: 1.7397 - val_acc: 0.8503 Epoch 21/37 6552/6680 [============================>.] - ETA: 0s - loss: 0.1927 - acc: 0.9750Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.1896 - acc: 0.9753 - val_loss: 1.6959 - val_acc: 0.8527 Epoch 22/37 6624/6680 [============================>.] - ETA: 0s - loss: 0.2141 - acc: 0.9740Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.2132 - acc: 0.9741 - val_loss: 1.7483 - val_acc: 0.8431 Epoch 23/37 6624/6680 [============================>.] - ETA: 0s - loss: 0.1734 - acc: 0.9766Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.1745 - acc: 0.9762 - val_loss: 1.8066 - val_acc: 0.8431 Epoch 24/37 6660/6680 [============================>.] - ETA: 0s - loss: 0.1942 - acc: 0.9740Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1936 - acc: 0.9741 - val_loss: 1.7419 - val_acc: 0.8527 Epoch 25/37 6588/6680 [============================>.] - ETA: 0s - loss: 0.2100 - acc: 0.9724Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.2097 - acc: 0.9725 - val_loss: 1.8478 - val_acc: 0.8443 Epoch 26/37 6624/6680 [============================>.] - ETA: 0s - loss: 0.2128 - acc: 0.9731Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 4s 541us/step - loss: 0.2112 - acc: 0.9732 - val_loss: 1.7536 - val_acc: 0.8383 Epoch 27/37 6588/6680 [============================>.] - ETA: 0s - loss: 0.2277 - acc: 0.9701Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.2324 - acc: 0.9699 - val_loss: 1.8146 - val_acc: 0.8455 Epoch 28/37 6660/6680 [============================>.] - ETA: 0s - loss: 0.1935 - acc: 0.9739Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.1961 - acc: 0.9737 - val_loss: 1.7656 - val_acc: 0.8479 Epoch 29/37 6660/6680 [============================>.] - ETA: 0s - loss: 0.2214 - acc: 0.9722Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.2208 - acc: 0.9723 - val_loss: 1.8015 - val_acc: 0.8431 Epoch 30/37 6660/6680 [============================>.] - ETA: 0s - loss: 0.2132 - acc: 0.9739Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.2126 - acc: 0.9740 - val_loss: 1.7189 - val_acc: 0.8587 Epoch 31/37 6660/6680 [============================>.] - ETA: 0s - loss: 0.2182 - acc: 0.9725Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.2176 - acc: 0.9726 - val_loss: 1.7318 - val_acc: 0.8527 Epoch 32/37 6588/6680 [============================>.] - ETA: 0s - loss: 0.2011 - acc: 0.9727Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.2034 - acc: 0.9726 - val_loss: 1.7111 - val_acc: 0.8599 Epoch 33/37 6588/6680 [============================>.] - ETA: 0s - loss: 0.1904 - acc: 0.9765Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1909 - acc: 0.9763 - val_loss: 1.8571 - val_acc: 0.8479 Epoch 34/37 6660/6680 [============================>.] - ETA: 0s - loss: 0.1981 - acc: 0.9757Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.1975 - acc: 0.9757 - val_loss: 1.8084 - val_acc: 0.8527 Epoch 35/37 6660/6680 [============================>.] - ETA: 0s - loss: 0.2103 - acc: 0.9730Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.2098 - acc: 0.9729 - val_loss: 1.7654 - val_acc: 0.8467 Epoch 36/37 6660/6680 [============================>.] - ETA: 0s - loss: 0.1870 - acc: 0.9779Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.1865 - acc: 0.9780 - val_loss: 1.7981 - val_acc: 0.8431 Epoch 37/37 6588/6680 [============================>.] - ETA: 0s - loss: 0.1937 - acc: 0.9768Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.1938 - acc: 0.9765 - val_loss: 1.7574 - val_acc: 0.8563 Batch size=36 Epoch=40 Train on 6680 samples, validate on 835 samples Epoch 1/40 6624/6680 [============================>.] - ETA: 0s - loss: 0.2235 - acc: 0.9681Epoch 00001: val_loss improved from inf to 1.71629, saving model to saved_models2/weights.best.ResNet_bs36_ep40.hdf5 6680/6680 [==============================] - 4s 554us/step - loss: 0.2234 - acc: 0.9681 - val_loss: 1.7163 - val_acc: 0.8527 Epoch 2/40 6660/6680 [============================>.] - ETA: 0s - loss: 0.2102 - acc: 0.9703Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 4s 542us/step - loss: 0.2111 - acc: 0.9701 - val_loss: 1.7657 - val_acc: 0.8503 Epoch 3/40 6660/6680 [============================>.] - ETA: 0s - loss: 0.2106 - acc: 0.9721Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.2113 - acc: 0.9720 - val_loss: 1.7317 - val_acc: 0.8515 Epoch 4/40 6660/6680 [============================>.] - ETA: 0s - loss: 0.2075 - acc: 0.9703Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.2068 - acc: 0.9704 - val_loss: 1.7726 - val_acc: 0.8479 Epoch 5/40 6588/6680 [============================>.] - ETA: 0s - loss: 0.1890 - acc: 0.9733Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.1922 - acc: 0.9731 - val_loss: 1.8228 - val_acc: 0.8395 Epoch 6/40 6588/6680 [============================>.] - ETA: 0s - loss: 0.2316 - acc: 0.9686Epoch 00006: val_loss improved from 1.71629 to 1.71116, saving model to saved_models2/weights.best.ResNet_bs36_ep40.hdf5 6680/6680 [==============================] - 4s 538us/step - loss: 0.2341 - acc: 0.9680 - val_loss: 1.7112 - val_acc: 0.8515 Epoch 7/40 6660/6680 [============================>.] - ETA: 0s - loss: 0.2016 - acc: 0.9724Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.2010 - acc: 0.9725 - val_loss: 1.7687 - val_acc: 0.8563 Epoch 8/40 6660/6680 [============================>.] - ETA: 0s - loss: 0.2039 - acc: 0.9734Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.2033 - acc: 0.9735 - val_loss: 1.7353 - val_acc: 0.8539 Epoch 9/40 6624/6680 [============================>.] - ETA: 0s - loss: 0.2333 - acc: 0.9715Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.2368 - acc: 0.9713 - val_loss: 1.7669 - val_acc: 0.8515 Epoch 10/40 6624/6680 [============================>.] - ETA: 0s - loss: 0.2147 - acc: 0.9707Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.2132 - acc: 0.9708 - val_loss: 1.7377 - val_acc: 0.8527 Epoch 11/40 6624/6680 [============================>.] - ETA: 0s - loss: 0.1958 - acc: 0.9746Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.1942 - acc: 0.9749 - val_loss: 1.7925 - val_acc: 0.8455 Epoch 12/40 6624/6680 [============================>.] - ETA: 0s - loss: 0.1916 - acc: 0.9746Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1943 - acc: 0.9744 - val_loss: 1.7652 - val_acc: 0.8503 Epoch 13/40 6624/6680 [============================>.] - ETA: 0s - loss: 0.2224 - acc: 0.9758Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.2215 - acc: 0.9759 - val_loss: 1.7391 - val_acc: 0.8539 Epoch 14/40 6624/6680 [============================>.] - ETA: 0s - loss: 0.2441 - acc: 0.9697Epoch 00014: val_loss improved from 1.71116 to 1.70034, saving model to saved_models2/weights.best.ResNet_bs36_ep40.hdf5 6680/6680 [==============================] - 4s 538us/step - loss: 0.2445 - acc: 0.9698 - val_loss: 1.7003 - val_acc: 0.8587 Epoch 15/40 6624/6680 [============================>.] - ETA: 0s - loss: 0.2023 - acc: 0.9740Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.2021 - acc: 0.9740 - val_loss: 1.7372 - val_acc: 0.8527 Epoch 16/40 6660/6680 [============================>.] - ETA: 0s - loss: 0.1969 - acc: 0.9745Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.1973 - acc: 0.9744 - val_loss: 1.7079 - val_acc: 0.8563 Epoch 17/40 6660/6680 [============================>.] - ETA: 0s - loss: 0.2109 - acc: 0.9712Epoch 00017: val_loss improved from 1.70034 to 1.67153, saving model to saved_models2/weights.best.ResNet_bs36_ep40.hdf5 6680/6680 [==============================] - 4s 555us/step - loss: 0.2103 - acc: 0.9713 - val_loss: 1.6715 - val_acc: 0.8575 Epoch 18/40 6660/6680 [============================>.] - ETA: 0s - loss: 0.1965 - acc: 0.9745Epoch 00018: val_loss improved from 1.67153 to 1.61047, saving model to saved_models2/weights.best.ResNet_bs36_ep40.hdf5 6680/6680 [==============================] - 4s 548us/step - loss: 0.1959 - acc: 0.9746 - val_loss: 1.6105 - val_acc: 0.8647 Epoch 19/40 6624/6680 [============================>.] - ETA: 0s - loss: 0.2098 - acc: 0.9710Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.2084 - acc: 0.9711 - val_loss: 1.7775 - val_acc: 0.8563 Epoch 20/40 6624/6680 [============================>.] - ETA: 0s - loss: 0.2061 - acc: 0.9737Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.2044 - acc: 0.9740 - val_loss: 1.7165 - val_acc: 0.8587 Epoch 21/40 6660/6680 [============================>.] - ETA: 0s - loss: 0.2317 - acc: 0.9710Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.2318 - acc: 0.9710 - val_loss: 1.6813 - val_acc: 0.8527 Epoch 22/40 6624/6680 [============================>.] - ETA: 0s - loss: 0.1893 - acc: 0.9739Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.1943 - acc: 0.9735 - val_loss: 1.6457 - val_acc: 0.8611 Epoch 23/40 6624/6680 [============================>.] - ETA: 0s - loss: 0.2286 - acc: 0.9721Epoch 00023: val_loss improved from 1.61047 to 1.60085, saving model to saved_models2/weights.best.ResNet_bs36_ep40.hdf5 6680/6680 [==============================] - 4s 548us/step - loss: 0.2268 - acc: 0.9723 - val_loss: 1.6008 - val_acc: 0.8671 Epoch 24/40 6624/6680 [============================>.] - ETA: 0s - loss: 0.2088 - acc: 0.9743Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 4s 544us/step - loss: 0.2080 - acc: 0.9744 - val_loss: 1.7435 - val_acc: 0.8587 Epoch 25/40 6660/6680 [============================>.] - ETA: 0s - loss: 0.1918 - acc: 0.9751Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.1923 - acc: 0.9750 - val_loss: 1.7605 - val_acc: 0.8623 Epoch 26/40 6660/6680 [============================>.] - ETA: 0s - loss: 0.1974 - acc: 0.9740Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1968 - acc: 0.9741 - val_loss: 1.8156 - val_acc: 0.8491 Epoch 27/40 6624/6680 [============================>.] - ETA: 0s - loss: 0.2090 - acc: 0.9739Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 4s 541us/step - loss: 0.2075 - acc: 0.9740 - val_loss: 1.7038 - val_acc: 0.8551 Epoch 28/40 6660/6680 [============================>.] - ETA: 0s - loss: 0.2203 - acc: 0.9740Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.2197 - acc: 0.9741 - val_loss: 1.7445 - val_acc: 0.8563 Epoch 29/40 6624/6680 [============================>.] - ETA: 0s - loss: 0.2387 - acc: 0.9710Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.2386 - acc: 0.9710 - val_loss: 1.7786 - val_acc: 0.8587 Epoch 30/40 6624/6680 [============================>.] - ETA: 0s - loss: 0.1789 - acc: 0.9764Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.1775 - acc: 0.9766 - val_loss: 1.8230 - val_acc: 0.8503 Epoch 31/40 6660/6680 [============================>.] - ETA: 0s - loss: 0.2164 - acc: 0.9725Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.2203 - acc: 0.9723 - val_loss: 1.7785 - val_acc: 0.8551 Epoch 32/40 6660/6680 [============================>.] - ETA: 0s - loss: 0.2354 - acc: 0.9710Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.2356 - acc: 0.9710 - val_loss: 1.8548 - val_acc: 0.8479 Epoch 33/40 6624/6680 [============================>.] - ETA: 0s - loss: 0.2163 - acc: 0.9743Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.2177 - acc: 0.9743 - val_loss: 1.8467 - val_acc: 0.8455 Epoch 34/40 6624/6680 [============================>.] - ETA: 0s - loss: 0.1890 - acc: 0.9745Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 4s 553us/step - loss: 0.1876 - acc: 0.9746 - val_loss: 1.8162 - val_acc: 0.8467 Epoch 35/40 6660/6680 [============================>.] - ETA: 0s - loss: 0.2042 - acc: 0.9760Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.2036 - acc: 0.9760 - val_loss: 1.8091 - val_acc: 0.8515 Epoch 36/40 6660/6680 [============================>.] - ETA: 0s - loss: 0.1854 - acc: 0.9760Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.1848 - acc: 0.9760 - val_loss: 1.7366 - val_acc: 0.8623 Epoch 37/40 6588/6680 [============================>.] - ETA: 0s - loss: 0.2249 - acc: 0.9715Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.2219 - acc: 0.9719 - val_loss: 1.7641 - val_acc: 0.8611 Epoch 38/40 6588/6680 [============================>.] - ETA: 0s - loss: 0.2288 - acc: 0.9737Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.2283 - acc: 0.9738 - val_loss: 1.8036 - val_acc: 0.8623 Epoch 39/40 6624/6680 [============================>.] - ETA: 0s - loss: 0.1913 - acc: 0.9745Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 4s 542us/step - loss: 0.1919 - acc: 0.9743 - val_loss: 1.7460 - val_acc: 0.8527 Epoch 40/40 6624/6680 [============================>.] - ETA: 0s - loss: 0.2100 - acc: 0.9731Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.2086 - acc: 0.9732 - val_loss: 1.8714 - val_acc: 0.8479 Batch size=36 Epoch=50 Train on 6680 samples, validate on 835 samples Epoch 1/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.2126 - acc: 0.9727Epoch 00001: val_loss improved from inf to 1.71068, saving model to saved_models2/weights.best.ResNet_bs36_ep50.hdf5 6680/6680 [==============================] - 4s 543us/step - loss: 0.2139 - acc: 0.9726 - val_loss: 1.7107 - val_acc: 0.8563 Epoch 2/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.2159 - acc: 0.9725Epoch 00002: val_loss improved from 1.71068 to 1.65420, saving model to saved_models2/weights.best.ResNet_bs36_ep50.hdf5 6680/6680 [==============================] - 4s 552us/step - loss: 0.2155 - acc: 0.9725 - val_loss: 1.6542 - val_acc: 0.8623 Epoch 3/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.1978 - acc: 0.9749Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 4s 547us/step - loss: 0.1981 - acc: 0.9750 - val_loss: 1.7233 - val_acc: 0.8455 Epoch 4/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.2238 - acc: 0.9721Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 4s 545us/step - loss: 0.2232 - acc: 0.9722 - val_loss: 1.7697 - val_acc: 0.8443 Epoch 5/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.2346 - acc: 0.9710Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 4s 541us/step - loss: 0.2329 - acc: 0.9711 - val_loss: 1.7041 - val_acc: 0.8551 Epoch 6/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.1976 - acc: 0.9740Epoch 00006: val_loss improved from 1.65420 to 1.64808, saving model to saved_models2/weights.best.ResNet_bs36_ep50.hdf5 6680/6680 [==============================] - 4s 541us/step - loss: 0.1972 - acc: 0.9740 - val_loss: 1.6481 - val_acc: 0.8503 Epoch 7/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.2554 - acc: 0.9694Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.2568 - acc: 0.9693 - val_loss: 1.6885 - val_acc: 0.8551 Epoch 8/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.1821 - acc: 0.9758Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.1821 - acc: 0.9757 - val_loss: 1.7016 - val_acc: 0.8539 Epoch 9/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.2097 - acc: 0.9731Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 4s 526us/step - loss: 0.2140 - acc: 0.9726 - val_loss: 1.6538 - val_acc: 0.8575 Epoch 10/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.2064 - acc: 0.9746Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.2059 - acc: 0.9747 - val_loss: 1.6961 - val_acc: 0.8527 Epoch 11/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.2245 - acc: 0.9733Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.2259 - acc: 0.9731 - val_loss: 1.6971 - val_acc: 0.8551 Epoch 12/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.2187 - acc: 0.9716Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.2231 - acc: 0.9714 - val_loss: 1.7222 - val_acc: 0.8599 Epoch 13/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.2124 - acc: 0.9736Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.2117 - acc: 0.9737 - val_loss: 1.6610 - val_acc: 0.8539 Epoch 14/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.2216 - acc: 0.9737Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.2234 - acc: 0.9737 - val_loss: 1.6563 - val_acc: 0.8575 Epoch 15/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.1825 - acc: 0.9789Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.1846 - acc: 0.9784 - val_loss: 1.6610 - val_acc: 0.8623 Epoch 16/50 6588/6680 [============================>.] - ETA: 0s - loss: 0.1876 - acc: 0.9747Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.1902 - acc: 0.9746 - val_loss: 1.6840 - val_acc: 0.8503 Epoch 17/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.2058 - acc: 0.9745Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.2065 - acc: 0.9746 - val_loss: 1.6973 - val_acc: 0.8563 Epoch 18/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.1903 - acc: 0.9743Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.1905 - acc: 0.9743 - val_loss: 1.6814 - val_acc: 0.8599 Epoch 19/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.2057 - acc: 0.9733Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.2051 - acc: 0.9734 - val_loss: 1.7473 - val_acc: 0.8575 Epoch 20/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.1712 - acc: 0.9783Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 4s 553us/step - loss: 0.1697 - acc: 0.9784 - val_loss: 1.6989 - val_acc: 0.8575 Epoch 21/50 6588/6680 [============================>.] - ETA: 0s - loss: 0.2091 - acc: 0.9721Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.2127 - acc: 0.9717 - val_loss: 1.7971 - val_acc: 0.8515 Epoch 22/50 6588/6680 [============================>.] - ETA: 0s - loss: 0.1998 - acc: 0.9763Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 3s 522us/step - loss: 0.2013 - acc: 0.9760 - val_loss: 1.8139 - val_acc: 0.8491 Epoch 23/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.2182 - acc: 0.9736Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.2189 - acc: 0.9735 - val_loss: 1.8076 - val_acc: 0.8551 Epoch 24/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.1991 - acc: 0.9758Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1985 - acc: 0.9759 - val_loss: 1.7946 - val_acc: 0.8575 Epoch 25/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.1716 - acc: 0.9787Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.1750 - acc: 0.9786 - val_loss: 1.7697 - val_acc: 0.8563 Epoch 26/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.2375 - acc: 0.9725Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.2379 - acc: 0.9726 - val_loss: 1.7972 - val_acc: 0.8539 Epoch 27/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.2090 - acc: 0.9748Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.2087 - acc: 0.9747 - val_loss: 1.7589 - val_acc: 0.8599 Epoch 28/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.1914 - acc: 0.9769Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 4s 549us/step - loss: 0.1929 - acc: 0.9768 - val_loss: 1.7564 - val_acc: 0.8575 Epoch 29/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.2066 - acc: 0.9745Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.2071 - acc: 0.9746 - val_loss: 1.8331 - val_acc: 0.8539 Epoch 30/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.2043 - acc: 0.9748Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.2026 - acc: 0.9750 - val_loss: 1.8770 - val_acc: 0.8551 Epoch 31/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.2036 - acc: 0.9751Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.2030 - acc: 0.9751 - val_loss: 1.9126 - val_acc: 0.8491 Epoch 32/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.2038 - acc: 0.9761Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.2048 - acc: 0.9760 - val_loss: 1.8715 - val_acc: 0.8527 Epoch 33/50 6588/6680 [============================>.] - ETA: 0s - loss: 0.2179 - acc: 0.9756Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.2166 - acc: 0.9754 - val_loss: 1.7915 - val_acc: 0.8611 Epoch 34/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.1994 - acc: 0.9763Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.1978 - acc: 0.9765 - val_loss: 1.8318 - val_acc: 0.8551 Epoch 35/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.2190 - acc: 0.9745Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.2202 - acc: 0.9744 - val_loss: 1.8119 - val_acc: 0.8527 Epoch 36/50 6588/6680 [============================>.] - ETA: 0s - loss: 0.1751 - acc: 0.9789Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1746 - acc: 0.9789 - val_loss: 1.7348 - val_acc: 0.8503 Epoch 37/50 6588/6680 [============================>.] - ETA: 0s - loss: 0.1734 - acc: 0.9795Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 4s 546us/step - loss: 0.1741 - acc: 0.9792 - val_loss: 1.8429 - val_acc: 0.8503 Epoch 38/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.1709 - acc: 0.9802Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.1715 - acc: 0.9799 - val_loss: 1.7520 - val_acc: 0.8587 Epoch 39/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.1778 - acc: 0.9778Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1773 - acc: 0.9778 - val_loss: 1.7298 - val_acc: 0.8575 Epoch 40/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.2034 - acc: 0.9746Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.2028 - acc: 0.9747 - val_loss: 1.7199 - val_acc: 0.8551 Epoch 41/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.2127 - acc: 0.9755Epoch 00041: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.2141 - acc: 0.9753 - val_loss: 1.7542 - val_acc: 0.8551 Epoch 42/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.1913 - acc: 0.9745Epoch 00042: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.1897 - acc: 0.9747 - val_loss: 1.7214 - val_acc: 0.8563 Epoch 43/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.2035 - acc: 0.9767Epoch 00043: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.2029 - acc: 0.9768 - val_loss: 1.8409 - val_acc: 0.8527 Epoch 44/50 6588/6680 [============================>.] - ETA: 0s - loss: 0.1722 - acc: 0.9803Epoch 00044: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.1728 - acc: 0.9799 - val_loss: 1.9316 - val_acc: 0.8515 Epoch 45/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.2135 - acc: 0.9746Epoch 00045: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.2129 - acc: 0.9747 - val_loss: 1.7724 - val_acc: 0.8527 Epoch 46/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.2072 - acc: 0.9754Epoch 00046: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.2054 - acc: 0.9756 - val_loss: 1.7570 - val_acc: 0.8575 Epoch 47/50 6660/6680 [============================>.] - ETA: 0s - loss: 0.1560 - acc: 0.9821Epoch 00047: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1578 - acc: 0.9819 - val_loss: 1.7658 - val_acc: 0.8575 Epoch 48/50 6588/6680 [============================>.] - ETA: 0s - loss: 0.1891 - acc: 0.9766Epoch 00048: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1940 - acc: 0.9762 - val_loss: 1.8177 - val_acc: 0.8599 Epoch 49/50 6588/6680 [============================>.] - ETA: 0s - loss: 0.1753 - acc: 0.9783Epoch 00049: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.1751 - acc: 0.9783 - val_loss: 1.8535 - val_acc: 0.8515 Epoch 50/50 6624/6680 [============================>.] - ETA: 0s - loss: 0.1928 - acc: 0.9772Epoch 00050: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1936 - acc: 0.9772 - val_loss: 1.7073 - val_acc: 0.8611 Batch size=36 Epoch=55 Train on 6680 samples, validate on 835 samples Epoch 1/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.2135 - acc: 0.9752Epoch 00001: val_loss improved from inf to 1.67392, saving model to saved_models2/weights.best.ResNet_bs36_ep55.hdf5 6680/6680 [==============================] - 4s 542us/step - loss: 0.2117 - acc: 0.9754 - val_loss: 1.6739 - val_acc: 0.8479 Epoch 2/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.2073 - acc: 0.9728Epoch 00002: val_loss improved from 1.67392 to 1.64500, saving model to saved_models2/weights.best.ResNet_bs36_ep55.hdf5 6680/6680 [==============================] - 4s 531us/step - loss: 0.2079 - acc: 0.9728 - val_loss: 1.6450 - val_acc: 0.8599 Epoch 3/55 6588/6680 [============================>.] - ETA: 0s - loss: 0.2006 - acc: 0.9737Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.1986 - acc: 0.9737 - val_loss: 1.7052 - val_acc: 0.8491 Epoch 4/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.2330 - acc: 0.9716Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 4s 541us/step - loss: 0.2323 - acc: 0.9717 - val_loss: 1.7535 - val_acc: 0.8479 Epoch 5/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.2159 - acc: 0.9737Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.2141 - acc: 0.9740 - val_loss: 1.6464 - val_acc: 0.8647 Epoch 6/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.1974 - acc: 0.9766Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 4s 545us/step - loss: 0.1982 - acc: 0.9766 - val_loss: 1.7226 - val_acc: 0.8599 Epoch 7/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.2147 - acc: 0.9736Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 4s 545us/step - loss: 0.2185 - acc: 0.9732 - val_loss: 1.6494 - val_acc: 0.8611 Epoch 8/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.1736 - acc: 0.9775Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.1739 - acc: 0.9774 - val_loss: 1.7370 - val_acc: 0.8515 Epoch 9/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.2062 - acc: 0.9739Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.2048 - acc: 0.9740 - val_loss: 1.7680 - val_acc: 0.8563 Epoch 10/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.1852 - acc: 0.9745Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.1836 - acc: 0.9747 - val_loss: 1.7212 - val_acc: 0.8587 Epoch 11/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.2253 - acc: 0.9748Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.2258 - acc: 0.9747 - val_loss: 1.7188 - val_acc: 0.8611 Epoch 12/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.1716 - acc: 0.9780Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.1717 - acc: 0.9778 - val_loss: 1.8003 - val_acc: 0.8599 Epoch 13/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.2205 - acc: 0.9736Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.2198 - acc: 0.9737 - val_loss: 1.8085 - val_acc: 0.8575 Epoch 14/55 6588/6680 [============================>.] - ETA: 0s - loss: 0.1919 - acc: 0.9742Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1893 - acc: 0.9746 - val_loss: 1.7948 - val_acc: 0.8563 Epoch 15/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.2101 - acc: 0.9734Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 4s 542us/step - loss: 0.2095 - acc: 0.9735 - val_loss: 1.7988 - val_acc: 0.8527 Epoch 16/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.2349 - acc: 0.9722Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.2342 - acc: 0.9723 - val_loss: 1.7683 - val_acc: 0.8551 Epoch 17/55 6588/6680 [============================>.] - ETA: 0s - loss: 0.1990 - acc: 0.9753Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.1987 - acc: 0.9754 - val_loss: 1.7568 - val_acc: 0.8515 Epoch 18/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.1829 - acc: 0.9770Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1824 - acc: 0.9771 - val_loss: 1.7447 - val_acc: 0.8527 Epoch 19/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.2366 - acc: 0.9737Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.2346 - acc: 0.9740 - val_loss: 1.7358 - val_acc: 0.8563 Epoch 20/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.1885 - acc: 0.9774Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 4s 541us/step - loss: 0.1881 - acc: 0.9774 - val_loss: 1.7690 - val_acc: 0.8527 Epoch 21/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.2053 - acc: 0.9751Epoch 00021: val_loss improved from 1.64500 to 1.63541, saving model to saved_models2/weights.best.ResNet_bs36_ep55.hdf5 6680/6680 [==============================] - 4s 542us/step - loss: 0.2043 - acc: 0.9751 - val_loss: 1.6354 - val_acc: 0.8647 Epoch 22/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.1734 - acc: 0.9790Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.1728 - acc: 0.9790 - val_loss: 1.7122 - val_acc: 0.8587 Epoch 23/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.1806 - acc: 0.9781Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 4s 542us/step - loss: 0.1800 - acc: 0.9781 - val_loss: 1.7285 - val_acc: 0.8527 Epoch 24/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.2007 - acc: 0.9746Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.2009 - acc: 0.9746 - val_loss: 1.7216 - val_acc: 0.8587 Epoch 25/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.2210 - acc: 0.9734Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.2200 - acc: 0.9735 - val_loss: 1.7650 - val_acc: 0.8551 Epoch 26/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.2249 - acc: 0.9752Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.2237 - acc: 0.9753 - val_loss: 1.7636 - val_acc: 0.8467 Epoch 27/55 6588/6680 [============================>.] - ETA: 0s - loss: 0.1789 - acc: 0.9766Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.1806 - acc: 0.9763 - val_loss: 1.7544 - val_acc: 0.8539 Epoch 28/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.1952 - acc: 0.9751Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1936 - acc: 0.9753 - val_loss: 1.8146 - val_acc: 0.8539 Epoch 29/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.1982 - acc: 0.9748Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.2000 - acc: 0.9747 - val_loss: 1.8471 - val_acc: 0.8515 Epoch 30/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.2085 - acc: 0.9751Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.2079 - acc: 0.9751 - val_loss: 1.8337 - val_acc: 0.8551 Epoch 31/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.2264 - acc: 0.9733Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.2258 - acc: 0.9734 - val_loss: 1.8259 - val_acc: 0.8527 Epoch 32/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.2042 - acc: 0.9774Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.2025 - acc: 0.9775 - val_loss: 1.8255 - val_acc: 0.8479 Epoch 33/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.1916 - acc: 0.9774Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.1932 - acc: 0.9772 - val_loss: 1.8402 - val_acc: 0.8515 Epoch 34/55 6588/6680 [============================>.] - ETA: 0s - loss: 0.1775 - acc: 0.9780Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.1810 - acc: 0.9778 - val_loss: 1.8115 - val_acc: 0.8503 Epoch 35/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.2073 - acc: 0.9749Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.2068 - acc: 0.9750 - val_loss: 1.8045 - val_acc: 0.8455 Epoch 36/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.2039 - acc: 0.9779Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.2033 - acc: 0.9780 - val_loss: 1.8855 - val_acc: 0.8491 Epoch 37/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.1867 - acc: 0.9766Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.1864 - acc: 0.9765 - val_loss: 1.8303 - val_acc: 0.8491 Epoch 38/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.2035 - acc: 0.9763Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.2072 - acc: 0.9760 - val_loss: 1.8129 - val_acc: 0.8479 Epoch 39/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.2119 - acc: 0.9760Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.2143 - acc: 0.9757 - val_loss: 1.8393 - val_acc: 0.8539 Epoch 40/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.2120 - acc: 0.9748Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.2130 - acc: 0.9746 - val_loss: 1.8798 - val_acc: 0.8467 Epoch 41/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.1995 - acc: 0.9766Epoch 00041: val_loss did not improve 6680/6680 [==============================] - 4s 544us/step - loss: 0.1989 - acc: 0.9766 - val_loss: 1.7576 - val_acc: 0.8515 Epoch 42/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.2089 - acc: 0.9780Epoch 00042: val_loss did not improve 6680/6680 [==============================] - 4s 542us/step - loss: 0.2072 - acc: 0.9781 - val_loss: 1.8129 - val_acc: 0.8503 Epoch 43/55 6588/6680 [============================>.] - ETA: 0s - loss: 0.1933 - acc: 0.9786Epoch 00043: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.1961 - acc: 0.9784 - val_loss: 1.7830 - val_acc: 0.8587 Epoch 44/55 6588/6680 [============================>.] - ETA: 0s - loss: 0.1946 - acc: 0.9778Epoch 00044: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1978 - acc: 0.9772 - val_loss: 1.7904 - val_acc: 0.8623 Epoch 45/55 6588/6680 [============================>.] - ETA: 0s - loss: 0.2206 - acc: 0.9743Epoch 00045: val_loss did not improve 6680/6680 [==============================] - 4s 541us/step - loss: 0.2180 - acc: 0.9746 - val_loss: 1.7979 - val_acc: 0.8551 Epoch 46/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.1649 - acc: 0.9797Epoch 00046: val_loss did not improve 6680/6680 [==============================] - 4s 544us/step - loss: 0.1644 - acc: 0.9798 - val_loss: 1.8667 - val_acc: 0.8503 Epoch 47/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.1632 - acc: 0.9796Epoch 00047: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1627 - acc: 0.9796 - val_loss: 1.8407 - val_acc: 0.8479 Epoch 48/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.2429 - acc: 0.9727Epoch 00048: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.2439 - acc: 0.9725 - val_loss: 1.8288 - val_acc: 0.8563 Epoch 49/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.1909 - acc: 0.9778Epoch 00049: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.1903 - acc: 0.9778 - val_loss: 1.8913 - val_acc: 0.8563 Epoch 50/55 6588/6680 [============================>.] - ETA: 0s - loss: 0.1906 - acc: 0.9771Epoch 00050: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.1931 - acc: 0.9769 - val_loss: 1.8775 - val_acc: 0.8455 Epoch 51/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.1922 - acc: 0.9793Epoch 00051: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.1914 - acc: 0.9793 - val_loss: 1.8322 - val_acc: 0.8539 Epoch 52/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.1898 - acc: 0.9755Epoch 00052: val_loss did not improve 6680/6680 [==============================] - 4s 541us/step - loss: 0.1916 - acc: 0.9754 - val_loss: 1.8367 - val_acc: 0.8551 Epoch 53/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.1772 - acc: 0.9802Epoch 00053: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.1759 - acc: 0.9802 - val_loss: 1.8580 - val_acc: 0.8527 Epoch 54/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.1846 - acc: 0.9778Epoch 00054: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.1831 - acc: 0.9780 - val_loss: 1.7985 - val_acc: 0.8623 Epoch 55/55 6624/6680 [============================>.] - ETA: 0s - loss: 0.2000 - acc: 0.9758Epoch 00055: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1991 - acc: 0.9759 - val_loss: 1.8406 - val_acc: 0.8551 Batch size=37 Epoch=35 Train on 6680 samples, validate on 835 samples Epoch 1/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.2130 - acc: 0.9751Epoch 00001: val_loss improved from inf to 1.70352, saving model to saved_models2/weights.best.ResNet_bs37_ep35.hdf5 6680/6680 [==============================] - 4s 539us/step - loss: 0.2125 - acc: 0.9751 - val_loss: 1.7035 - val_acc: 0.8575 Epoch 2/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.1865 - acc: 0.9756Epoch 00002: val_loss improved from 1.70352 to 1.66106, saving model to saved_models2/weights.best.ResNet_bs37_ep35.hdf5 6680/6680 [==============================] - 4s 535us/step - loss: 0.1848 - acc: 0.9757 - val_loss: 1.6611 - val_acc: 0.8539 Epoch 3/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.2134 - acc: 0.9756Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.2112 - acc: 0.9756 - val_loss: 1.8065 - val_acc: 0.8611 Epoch 4/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.2337 - acc: 0.9713Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.2347 - acc: 0.9713 - val_loss: 1.7334 - val_acc: 0.8587 Epoch 5/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.2139 - acc: 0.9746Epoch 00005: val_loss improved from 1.66106 to 1.65937, saving model to saved_models2/weights.best.ResNet_bs37_ep35.hdf5 6680/6680 [==============================] - 4s 536us/step - loss: 0.2117 - acc: 0.9749 - val_loss: 1.6594 - val_acc: 0.8623 Epoch 6/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.2215 - acc: 0.9759Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.2210 - acc: 0.9759 - val_loss: 1.6858 - val_acc: 0.8623 Epoch 7/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.1714 - acc: 0.9784Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.1690 - acc: 0.9787 - val_loss: 1.7468 - val_acc: 0.8587 Epoch 8/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.1804 - acc: 0.9786Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.1841 - acc: 0.9783 - val_loss: 1.7452 - val_acc: 0.8623 Epoch 9/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.1627 - acc: 0.9790Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.1604 - acc: 0.9793 - val_loss: 1.7622 - val_acc: 0.8611 Epoch 10/35 6623/6680 [============================>.] - ETA: 0s - loss: 0.1939 - acc: 0.9761Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.1953 - acc: 0.9760 - val_loss: 1.7538 - val_acc: 0.8599 Epoch 11/35 6623/6680 [============================>.] - ETA: 0s - loss: 0.1736 - acc: 0.9787Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 4s 546us/step - loss: 0.1727 - acc: 0.9787 - val_loss: 1.7812 - val_acc: 0.8551 Epoch 12/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.1983 - acc: 0.9780Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.1990 - acc: 0.9778 - val_loss: 1.8289 - val_acc: 0.8515 Epoch 13/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.2105 - acc: 0.9743Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.2082 - acc: 0.9746 - val_loss: 1.8180 - val_acc: 0.8563 Epoch 14/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.2003 - acc: 0.9757Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.1981 - acc: 0.9759 - val_loss: 1.7166 - val_acc: 0.8611 Epoch 15/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.1776 - acc: 0.9775Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 4s 526us/step - loss: 0.1797 - acc: 0.9772 - val_loss: 1.8218 - val_acc: 0.8599 Epoch 16/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.2206 - acc: 0.9737Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 4s 526us/step - loss: 0.2194 - acc: 0.9740 - val_loss: 1.7906 - val_acc: 0.8539 Epoch 17/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.1827 - acc: 0.9778Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 3s 523us/step - loss: 0.1820 - acc: 0.9778 - val_loss: 1.7174 - val_acc: 0.8647 Epoch 18/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.1657 - acc: 0.9787Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 4s 524us/step - loss: 0.1679 - acc: 0.9787 - val_loss: 1.8582 - val_acc: 0.8551 Epoch 19/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.1816 - acc: 0.9784Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.1813 - acc: 0.9784 - val_loss: 1.8112 - val_acc: 0.8491 Epoch 20/35 6623/6680 [============================>.] - ETA: 0s - loss: 0.1844 - acc: 0.9760Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.1829 - acc: 0.9762 - val_loss: 1.8937 - val_acc: 0.8515 Epoch 21/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.1965 - acc: 0.9772Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.1977 - acc: 0.9771 - val_loss: 1.8383 - val_acc: 0.8491 Epoch 22/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.2023 - acc: 0.9753Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.2022 - acc: 0.9753 - val_loss: 1.7561 - val_acc: 0.8563 Epoch 23/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.1705 - acc: 0.9786Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.1698 - acc: 0.9787 - val_loss: 1.7440 - val_acc: 0.8611 Epoch 24/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.1881 - acc: 0.9784Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 3s 522us/step - loss: 0.1911 - acc: 0.9783 - val_loss: 1.7081 - val_acc: 0.8563 Epoch 25/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.1845 - acc: 0.9781Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.1864 - acc: 0.9780 - val_loss: 1.8112 - val_acc: 0.8539 Epoch 26/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.1894 - acc: 0.9762Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.1882 - acc: 0.9760 - val_loss: 1.7800 - val_acc: 0.8551 Epoch 27/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.2103 - acc: 0.9763Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 4s 526us/step - loss: 0.2099 - acc: 0.9763 - val_loss: 1.8006 - val_acc: 0.8599 Epoch 28/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.1875 - acc: 0.9783Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 4s 546us/step - loss: 0.1908 - acc: 0.9781 - val_loss: 1.7468 - val_acc: 0.8563 Epoch 29/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.1997 - acc: 0.9778Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.2036 - acc: 0.9777 - val_loss: 1.7157 - val_acc: 0.8599 Epoch 30/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.2162 - acc: 0.9748Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.2144 - acc: 0.9749 - val_loss: 1.8250 - val_acc: 0.8551 Epoch 31/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.1434 - acc: 0.9810Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.1415 - acc: 0.9813 - val_loss: 1.8280 - val_acc: 0.8539 Epoch 32/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.1942 - acc: 0.9774Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.1922 - acc: 0.9774 - val_loss: 1.8280 - val_acc: 0.8539 Epoch 33/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.1857 - acc: 0.9780Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 3s 522us/step - loss: 0.1843 - acc: 0.9781 - val_loss: 1.8910 - val_acc: 0.8527 Epoch 34/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.1763 - acc: 0.9801Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.1749 - acc: 0.9802 - val_loss: 1.8058 - val_acc: 0.8611 Epoch 35/35 6586/6680 [============================>.] - ETA: 0s - loss: 0.1710 - acc: 0.9790Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1697 - acc: 0.9790 - val_loss: 1.8984 - val_acc: 0.8455 Batch size=37 Epoch=37 Train on 6680 samples, validate on 835 samples Epoch 1/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1777 - acc: 0.9769Epoch 00001: val_loss improved from inf to 1.68524, saving model to saved_models2/weights.best.ResNet_bs37_ep37.hdf5 6680/6680 [==============================] - 4s 539us/step - loss: 0.1755 - acc: 0.9771 - val_loss: 1.6852 - val_acc: 0.8623 Epoch 2/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1986 - acc: 0.9756Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1984 - acc: 0.9756 - val_loss: 1.7902 - val_acc: 0.8575 Epoch 3/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1893 - acc: 0.9780Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.1876 - acc: 0.9781 - val_loss: 1.7333 - val_acc: 0.8647 Epoch 4/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1749 - acc: 0.9809Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.1744 - acc: 0.9804 - val_loss: 1.6864 - val_acc: 0.8575 Epoch 5/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1842 - acc: 0.9771Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1860 - acc: 0.9769 - val_loss: 1.7823 - val_acc: 0.8539 Epoch 6/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.2208 - acc: 0.9727Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.2261 - acc: 0.9723 - val_loss: 1.7973 - val_acc: 0.8587 Epoch 7/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1807 - acc: 0.9784Epoch 00007: val_loss improved from 1.68524 to 1.66562, saving model to saved_models2/weights.best.ResNet_bs37_ep37.hdf5 6680/6680 [==============================] - 4s 536us/step - loss: 0.1835 - acc: 0.9783 - val_loss: 1.6656 - val_acc: 0.8611 Epoch 8/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.2002 - acc: 0.9765Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.1983 - acc: 0.9766 - val_loss: 1.6746 - val_acc: 0.8611 Epoch 9/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1827 - acc: 0.9784Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.1810 - acc: 0.9784 - val_loss: 1.7670 - val_acc: 0.8563 Epoch 10/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.2095 - acc: 0.9746Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 3s 524us/step - loss: 0.2080 - acc: 0.9749 - val_loss: 1.8001 - val_acc: 0.8551 Epoch 11/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1960 - acc: 0.9777Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.1961 - acc: 0.9775 - val_loss: 1.8447 - val_acc: 0.8551 Epoch 12/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1903 - acc: 0.9772Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.1877 - acc: 0.9775 - val_loss: 1.7268 - val_acc: 0.8671 Epoch 13/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1688 - acc: 0.9792Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.1681 - acc: 0.9793 - val_loss: 1.7294 - val_acc: 0.8623 Epoch 14/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1943 - acc: 0.9766Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.1916 - acc: 0.9769 - val_loss: 1.7526 - val_acc: 0.8623 Epoch 15/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1733 - acc: 0.9790Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.1709 - acc: 0.9793 - val_loss: 1.7672 - val_acc: 0.8551 Epoch 16/37 6623/6680 [============================>.] - ETA: 0s - loss: 0.1679 - acc: 0.9804Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.1694 - acc: 0.9802 - val_loss: 1.7865 - val_acc: 0.8563 Epoch 17/37 6623/6680 [============================>.] - ETA: 0s - loss: 0.1861 - acc: 0.9799Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.1845 - acc: 0.9801 - val_loss: 1.6916 - val_acc: 0.8623 Epoch 18/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1638 - acc: 0.9803Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.1690 - acc: 0.9799 - val_loss: 1.7159 - val_acc: 0.8623 Epoch 19/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1865 - acc: 0.9766Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.1873 - acc: 0.9765 - val_loss: 1.7846 - val_acc: 0.8599 Epoch 20/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1781 - acc: 0.9775Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.1773 - acc: 0.9777 - val_loss: 1.8404 - val_acc: 0.8539 Epoch 21/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1692 - acc: 0.9789Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.1685 - acc: 0.9790 - val_loss: 1.8126 - val_acc: 0.8587 Epoch 22/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.2104 - acc: 0.9757Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.2108 - acc: 0.9756 - val_loss: 1.8845 - val_acc: 0.8563 Epoch 23/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1774 - acc: 0.9780Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.1773 - acc: 0.9781 - val_loss: 1.8281 - val_acc: 0.8527 Epoch 24/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1810 - acc: 0.9786Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.1839 - acc: 0.9783 - val_loss: 1.8551 - val_acc: 0.8551 Epoch 25/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1554 - acc: 0.9821Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.1571 - acc: 0.9819 - val_loss: 1.8344 - val_acc: 0.8515 Epoch 26/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.2021 - acc: 0.9756Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.2010 - acc: 0.9756 - val_loss: 1.7498 - val_acc: 0.8623 Epoch 27/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1883 - acc: 0.9787Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 4s 524us/step - loss: 0.1905 - acc: 0.9787 - val_loss: 1.7987 - val_acc: 0.8575 Epoch 28/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1737 - acc: 0.9795Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.1716 - acc: 0.9796 - val_loss: 1.7956 - val_acc: 0.8575 Epoch 29/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1563 - acc: 0.9806Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.1555 - acc: 0.9805 - val_loss: 1.7825 - val_acc: 0.8551 Epoch 30/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1505 - acc: 0.9821Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.1512 - acc: 0.9819 - val_loss: 1.8277 - val_acc: 0.8623 Epoch 31/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1634 - acc: 0.9822Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.1636 - acc: 0.9823 - val_loss: 1.9087 - val_acc: 0.8491 Epoch 32/37 6623/6680 [============================>.] - ETA: 0s - loss: 0.1976 - acc: 0.9770Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.1982 - acc: 0.9769 - val_loss: 1.7628 - val_acc: 0.8647 Epoch 33/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1692 - acc: 0.9815Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.1668 - acc: 0.9817 - val_loss: 1.7996 - val_acc: 0.8563 Epoch 34/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1612 - acc: 0.9822Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.1602 - acc: 0.9820 - val_loss: 1.8442 - val_acc: 0.8515 Epoch 35/37 6623/6680 [============================>.] - ETA: 0s - loss: 0.1769 - acc: 0.9802Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 3s 518us/step - loss: 0.1754 - acc: 0.9804 - val_loss: 1.8948 - val_acc: 0.8467 Epoch 36/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.2006 - acc: 0.9775Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.1979 - acc: 0.9778 - val_loss: 1.8400 - val_acc: 0.8563 Epoch 37/37 6586/6680 [============================>.] - ETA: 0s - loss: 0.1799 - acc: 0.9800Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.1799 - acc: 0.9799 - val_loss: 1.8637 - val_acc: 0.8587 Batch size=37 Epoch=40 Train on 6680 samples, validate on 835 samples Epoch 1/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1981 - acc: 0.9774Epoch 00001: val_loss improved from inf to 1.75257, saving model to saved_models2/weights.best.ResNet_bs37_ep40.hdf5 6680/6680 [==============================] - 4s 542us/step - loss: 0.1958 - acc: 0.9775 - val_loss: 1.7526 - val_acc: 0.8575 Epoch 2/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1781 - acc: 0.9774Epoch 00002: val_loss improved from 1.75257 to 1.66336, saving model to saved_models2/weights.best.ResNet_bs37_ep40.hdf5 6680/6680 [==============================] - 4s 535us/step - loss: 0.1756 - acc: 0.9777 - val_loss: 1.6634 - val_acc: 0.8647 Epoch 3/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1990 - acc: 0.9769Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.2010 - acc: 0.9768 - val_loss: 1.6720 - val_acc: 0.8635 Epoch 4/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1800 - acc: 0.9784Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 4s 526us/step - loss: 0.1799 - acc: 0.9786 - val_loss: 1.7363 - val_acc: 0.8647 Epoch 5/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.2098 - acc: 0.9763Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.2092 - acc: 0.9763 - val_loss: 1.7968 - val_acc: 0.8515 Epoch 6/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1844 - acc: 0.9768Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.1830 - acc: 0.9769 - val_loss: 1.7606 - val_acc: 0.8635 Epoch 7/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1528 - acc: 0.9816Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.1542 - acc: 0.9816 - val_loss: 1.8908 - val_acc: 0.8575 Epoch 8/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.2273 - acc: 0.9754Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.2292 - acc: 0.9753 - val_loss: 1.8824 - val_acc: 0.8587 Epoch 9/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1770 - acc: 0.9783Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.1746 - acc: 0.9786 - val_loss: 1.7654 - val_acc: 0.8515 Epoch 10/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.2009 - acc: 0.9751Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.2002 - acc: 0.9753 - val_loss: 1.8786 - val_acc: 0.8575 Epoch 11/40 6623/6680 [============================>.] - ETA: 0s - loss: 0.1737 - acc: 0.9786Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.1801 - acc: 0.9778 - val_loss: 1.8096 - val_acc: 0.8551 Epoch 12/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1776 - acc: 0.9790Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 4s 526us/step - loss: 0.1794 - acc: 0.9789 - val_loss: 1.8066 - val_acc: 0.8599 Epoch 13/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1695 - acc: 0.9780Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 4s 526us/step - loss: 0.1705 - acc: 0.9777 - val_loss: 1.8488 - val_acc: 0.8611 Epoch 14/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1881 - acc: 0.9783Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 3s 520us/step - loss: 0.1876 - acc: 0.9783 - val_loss: 1.8635 - val_acc: 0.8551 Epoch 15/40 6623/6680 [============================>.] - ETA: 0s - loss: 0.2063 - acc: 0.9778Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 3s 521us/step - loss: 0.2056 - acc: 0.9778 - val_loss: 1.8145 - val_acc: 0.8539 Epoch 16/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1852 - acc: 0.9787Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.1844 - acc: 0.9789 - val_loss: 1.8339 - val_acc: 0.8539 Epoch 17/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1905 - acc: 0.9760Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.1878 - acc: 0.9763 - val_loss: 1.8363 - val_acc: 0.8479 Epoch 18/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1935 - acc: 0.9766Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.1918 - acc: 0.9766 - val_loss: 1.8393 - val_acc: 0.8563 Epoch 19/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1856 - acc: 0.9781Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.1861 - acc: 0.9781 - val_loss: 1.8784 - val_acc: 0.8491 Epoch 20/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1755 - acc: 0.9800Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.1737 - acc: 0.9801 - val_loss: 1.8783 - val_acc: 0.8527 Epoch 21/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1756 - acc: 0.9798Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 3s 521us/step - loss: 0.1755 - acc: 0.9799 - val_loss: 1.8916 - val_acc: 0.8491 Epoch 22/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1855 - acc: 0.9786Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 4s 524us/step - loss: 0.1844 - acc: 0.9786 - val_loss: 2.0117 - val_acc: 0.8443 Epoch 23/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1859 - acc: 0.9801Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.1881 - acc: 0.9799 - val_loss: 1.9864 - val_acc: 0.8479 Epoch 24/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1815 - acc: 0.9787Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.1835 - acc: 0.9787 - val_loss: 1.9041 - val_acc: 0.8443 Epoch 25/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1976 - acc: 0.9777Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.1957 - acc: 0.9777 - val_loss: 1.8921 - val_acc: 0.8515 Epoch 26/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.2192 - acc: 0.9756Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.2168 - acc: 0.9757 - val_loss: 1.8800 - val_acc: 0.8599 Epoch 27/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1785 - acc: 0.9778Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.1796 - acc: 0.9777 - val_loss: 1.8814 - val_acc: 0.8527 Epoch 28/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1583 - acc: 0.9821Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.1561 - acc: 0.9823 - val_loss: 1.8202 - val_acc: 0.8575 Epoch 29/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.2063 - acc: 0.9771Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.2034 - acc: 0.9774 - val_loss: 1.9393 - val_acc: 0.8479 Epoch 30/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1574 - acc: 0.9807Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.1609 - acc: 0.9805 - val_loss: 1.9000 - val_acc: 0.8491 Epoch 31/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.2134 - acc: 0.9763Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.2160 - acc: 0.9757 - val_loss: 1.9178 - val_acc: 0.8491 Epoch 32/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1707 - acc: 0.9804Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 4s 526us/step - loss: 0.1699 - acc: 0.9802 - val_loss: 1.8310 - val_acc: 0.8587 Epoch 33/40 6623/6680 [============================>.] - ETA: 0s - loss: 0.1745 - acc: 0.9802Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 4s 526us/step - loss: 0.1736 - acc: 0.9802 - val_loss: 1.9120 - val_acc: 0.8539 Epoch 34/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1601 - acc: 0.9824Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.1594 - acc: 0.9823 - val_loss: 1.8268 - val_acc: 0.8527 Epoch 35/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.2096 - acc: 0.9769Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 3s 523us/step - loss: 0.2086 - acc: 0.9771 - val_loss: 1.7704 - val_acc: 0.8539 Epoch 36/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1699 - acc: 0.9797Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.1683 - acc: 0.9798 - val_loss: 1.8887 - val_acc: 0.8515 Epoch 37/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1637 - acc: 0.9803Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 4s 543us/step - loss: 0.1680 - acc: 0.9799 - val_loss: 1.8853 - val_acc: 0.8515 Epoch 38/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1910 - acc: 0.9786Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.1931 - acc: 0.9783 - val_loss: 1.8438 - val_acc: 0.8467 Epoch 39/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1674 - acc: 0.9813Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.1676 - acc: 0.9813 - val_loss: 1.8933 - val_acc: 0.8383 Epoch 40/40 6586/6680 [============================>.] - ETA: 0s - loss: 0.1886 - acc: 0.9790Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.1929 - acc: 0.9787 - val_loss: 1.8735 - val_acc: 0.8407 Batch size=37 Epoch=50 Train on 6680 samples, validate on 835 samples Epoch 1/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1980 - acc: 0.9757Epoch 00001: val_loss improved from inf to 1.59582, saving model to saved_models2/weights.best.ResNet_bs37_ep50.hdf5 6680/6680 [==============================] - 4s 531us/step - loss: 0.1952 - acc: 0.9760 - val_loss: 1.5958 - val_acc: 0.8707 Epoch 2/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1780 - acc: 0.9772Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.1771 - acc: 0.9772 - val_loss: 1.6889 - val_acc: 0.8599 Epoch 3/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1879 - acc: 0.9772Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.1870 - acc: 0.9774 - val_loss: 1.7193 - val_acc: 0.8467 Epoch 4/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.2121 - acc: 0.9751Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.2104 - acc: 0.9750 - val_loss: 1.8032 - val_acc: 0.8503 Epoch 5/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1867 - acc: 0.9803Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.1876 - acc: 0.9802 - val_loss: 1.8706 - val_acc: 0.8443 Epoch 6/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1768 - acc: 0.9786Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.1765 - acc: 0.9786 - val_loss: 1.8190 - val_acc: 0.8491 Epoch 7/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.2137 - acc: 0.9757Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.2154 - acc: 0.9754 - val_loss: 1.8598 - val_acc: 0.8467 Epoch 8/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.2105 - acc: 0.9756Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.2094 - acc: 0.9757 - val_loss: 1.8423 - val_acc: 0.8539 Epoch 9/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.2016 - acc: 0.9775Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.1995 - acc: 0.9777 - val_loss: 1.7886 - val_acc: 0.8527 Epoch 10/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1819 - acc: 0.9777Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 4s 537us/step - loss: 0.1848 - acc: 0.9774 - val_loss: 1.8631 - val_acc: 0.8479 Epoch 11/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1585 - acc: 0.9798Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.1602 - acc: 0.9796 - val_loss: 1.9001 - val_acc: 0.8515 Epoch 12/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1716 - acc: 0.9795Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.1723 - acc: 0.9793 - val_loss: 1.9215 - val_acc: 0.8491 Epoch 13/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1914 - acc: 0.9763Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.1920 - acc: 0.9763 - val_loss: 1.8247 - val_acc: 0.8551 Epoch 14/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1762 - acc: 0.9787Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 4s 526us/step - loss: 0.1811 - acc: 0.9784 - val_loss: 1.8624 - val_acc: 0.8467 Epoch 15/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.2015 - acc: 0.9766Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.2017 - acc: 0.9765 - val_loss: 1.8567 - val_acc: 0.8515 Epoch 16/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1923 - acc: 0.9783Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 4s 526us/step - loss: 0.1963 - acc: 0.9780 - val_loss: 1.9254 - val_acc: 0.8467 Epoch 17/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1400 - acc: 0.9813Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 3s 522us/step - loss: 0.1451 - acc: 0.9808 - val_loss: 1.8947 - val_acc: 0.8467 Epoch 18/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1932 - acc: 0.9777Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 3s 522us/step - loss: 0.1907 - acc: 0.9778 - val_loss: 1.7984 - val_acc: 0.8551 Epoch 19/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1864 - acc: 0.9797Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.1898 - acc: 0.9792 - val_loss: 1.8073 - val_acc: 0.8539 Epoch 20/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1877 - acc: 0.9781Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 3s 523us/step - loss: 0.1862 - acc: 0.9780 - val_loss: 1.8544 - val_acc: 0.8551 Epoch 21/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1724 - acc: 0.9800Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 3s 523us/step - loss: 0.1715 - acc: 0.9799 - val_loss: 1.8036 - val_acc: 0.8503 Epoch 22/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1694 - acc: 0.9783Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.1697 - acc: 0.9783 - val_loss: 1.7974 - val_acc: 0.8587 Epoch 23/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1892 - acc: 0.9790Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 3s 524us/step - loss: 0.1886 - acc: 0.9792 - val_loss: 1.8484 - val_acc: 0.8551 Epoch 24/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.2272 - acc: 0.9757Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 3s 520us/step - loss: 0.2295 - acc: 0.9756 - val_loss: 1.7813 - val_acc: 0.8539 Epoch 25/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1793 - acc: 0.9783Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 4s 526us/step - loss: 0.1782 - acc: 0.9784 - val_loss: 1.7815 - val_acc: 0.8527 Epoch 26/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1830 - acc: 0.9772Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 4s 526us/step - loss: 0.1851 - acc: 0.9771 - val_loss: 1.8443 - val_acc: 0.8563 Epoch 27/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1758 - acc: 0.9804Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.1780 - acc: 0.9801 - val_loss: 1.8617 - val_acc: 0.8491 Epoch 28/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1728 - acc: 0.9803Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1755 - acc: 0.9801 - val_loss: 1.9455 - val_acc: 0.8467 Epoch 29/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1916 - acc: 0.9777Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 4s 542us/step - loss: 0.1936 - acc: 0.9775 - val_loss: 1.9750 - val_acc: 0.8431 Epoch 30/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.2130 - acc: 0.9763Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.2107 - acc: 0.9763 - val_loss: 1.8778 - val_acc: 0.8551 Epoch 31/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1862 - acc: 0.9786Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.1847 - acc: 0.9786 - val_loss: 1.8579 - val_acc: 0.8527 Epoch 32/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1706 - acc: 0.9797Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.1730 - acc: 0.9795 - val_loss: 1.8197 - val_acc: 0.8575 Epoch 33/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1976 - acc: 0.9781Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.1960 - acc: 0.9783 - val_loss: 1.8251 - val_acc: 0.8515 Epoch 34/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1569 - acc: 0.9813Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.1601 - acc: 0.9811 - val_loss: 1.8400 - val_acc: 0.8563 Epoch 35/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.2005 - acc: 0.9781Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.2014 - acc: 0.9781 - val_loss: 1.8667 - val_acc: 0.8515 Epoch 36/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1713 - acc: 0.9809Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.1721 - acc: 0.9805 - val_loss: 1.8349 - val_acc: 0.8587 Epoch 37/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1685 - acc: 0.9798Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.1671 - acc: 0.9798 - val_loss: 1.7944 - val_acc: 0.8551 Epoch 38/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1839 - acc: 0.9798Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.1814 - acc: 0.9801 - val_loss: 1.8097 - val_acc: 0.8575 Epoch 39/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.2037 - acc: 0.9792Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.2023 - acc: 0.9793 - val_loss: 1.8409 - val_acc: 0.8515 Epoch 40/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1603 - acc: 0.9822Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.1600 - acc: 0.9822 - val_loss: 1.7496 - val_acc: 0.8575 Epoch 41/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1689 - acc: 0.9807Epoch 00041: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.1677 - acc: 0.9808 - val_loss: 1.7896 - val_acc: 0.8539 Epoch 42/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1821 - acc: 0.9790Epoch 00042: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.1831 - acc: 0.9790 - val_loss: 1.8020 - val_acc: 0.8563 Epoch 43/50 6623/6680 [============================>.] - ETA: 0s - loss: 0.1937 - acc: 0.9801Epoch 00043: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.1921 - acc: 0.9802 - val_loss: 1.7963 - val_acc: 0.8551 Epoch 44/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1810 - acc: 0.9807Epoch 00044: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.1802 - acc: 0.9808 - val_loss: 1.8630 - val_acc: 0.8527 Epoch 45/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1998 - acc: 0.9792Epoch 00045: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.1994 - acc: 0.9789 - val_loss: 1.8377 - val_acc: 0.8503 Epoch 46/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1536 - acc: 0.9813Epoch 00046: val_loss did not improve 6680/6680 [==============================] - 4s 526us/step - loss: 0.1596 - acc: 0.9805 - val_loss: 1.8614 - val_acc: 0.8551 Epoch 47/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1948 - acc: 0.9778Epoch 00047: val_loss did not improve 6680/6680 [==============================] - 4s 526us/step - loss: 0.1943 - acc: 0.9777 - val_loss: 1.8870 - val_acc: 0.8491 Epoch 48/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1837 - acc: 0.9787Epoch 00048: val_loss did not improve 6680/6680 [==============================] - 4s 524us/step - loss: 0.1819 - acc: 0.9787 - val_loss: 1.8924 - val_acc: 0.8515 Epoch 49/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1787 - acc: 0.9790Epoch 00049: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.1771 - acc: 0.9792 - val_loss: 1.9154 - val_acc: 0.8515 Epoch 50/50 6586/6680 [============================>.] - ETA: 0s - loss: 0.1846 - acc: 0.9772Epoch 00050: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.1820 - acc: 0.9775 - val_loss: 1.8670 - val_acc: 0.8515 Batch size=37 Epoch=55 Train on 6680 samples, validate on 835 samples Epoch 1/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1887 - acc: 0.9794Epoch 00001: val_loss improved from inf to 1.70679, saving model to saved_models2/weights.best.ResNet_bs37_ep55.hdf5 6680/6680 [==============================] - 4s 533us/step - loss: 0.1929 - acc: 0.9786 - val_loss: 1.7068 - val_acc: 0.8563 Epoch 2/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1593 - acc: 0.9798Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 3s 522us/step - loss: 0.1600 - acc: 0.9798 - val_loss: 1.7343 - val_acc: 0.8539 Epoch 3/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1624 - acc: 0.9801Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 3s 523us/step - loss: 0.1649 - acc: 0.9801 - val_loss: 1.7073 - val_acc: 0.8611 Epoch 4/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.2063 - acc: 0.9769Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.2064 - acc: 0.9769 - val_loss: 1.7214 - val_acc: 0.8575 Epoch 5/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1918 - acc: 0.9775Epoch 00005: val_loss improved from 1.70679 to 1.64211, saving model to saved_models2/weights.best.ResNet_bs37_ep55.hdf5 6680/6680 [==============================] - 4s 534us/step - loss: 0.1907 - acc: 0.9777 - val_loss: 1.6421 - val_acc: 0.8647 Epoch 6/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1771 - acc: 0.9781Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.1819 - acc: 0.9778 - val_loss: 1.6608 - val_acc: 0.8599 Epoch 7/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1619 - acc: 0.9827Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.1618 - acc: 0.9822 - val_loss: 1.7832 - val_acc: 0.8575 Epoch 8/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1802 - acc: 0.9795Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 3s 521us/step - loss: 0.1806 - acc: 0.9792 - val_loss: 1.6746 - val_acc: 0.8551 Epoch 9/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1680 - acc: 0.9798Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.1679 - acc: 0.9798 - val_loss: 1.7220 - val_acc: 0.8563 Epoch 10/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1969 - acc: 0.9760Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.1984 - acc: 0.9759 - val_loss: 1.7086 - val_acc: 0.8563 Epoch 11/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1810 - acc: 0.9783Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 4s 526us/step - loss: 0.1846 - acc: 0.9780 - val_loss: 1.7302 - val_acc: 0.8563 Epoch 12/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1880 - acc: 0.9772Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 4s 524us/step - loss: 0.1891 - acc: 0.9771 - val_loss: 1.6585 - val_acc: 0.8623 Epoch 13/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.2039 - acc: 0.9760Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 4s 526us/step - loss: 0.2052 - acc: 0.9757 - val_loss: 1.6768 - val_acc: 0.8671 Epoch 14/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1983 - acc: 0.9762Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 3s 523us/step - loss: 0.1955 - acc: 0.9765 - val_loss: 1.7455 - val_acc: 0.8587 Epoch 15/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1758 - acc: 0.9778Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.1777 - acc: 0.9777 - val_loss: 1.7022 - val_acc: 0.8695 Epoch 16/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1725 - acc: 0.9789Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.1728 - acc: 0.9787 - val_loss: 1.7201 - val_acc: 0.8659 Epoch 17/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1611 - acc: 0.9797Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 4s 527us/step - loss: 0.1593 - acc: 0.9798 - val_loss: 1.7958 - val_acc: 0.8491 Epoch 18/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1863 - acc: 0.9778Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.1847 - acc: 0.9780 - val_loss: 1.8531 - val_acc: 0.8491 Epoch 19/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.2136 - acc: 0.9775Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.2150 - acc: 0.9775 - val_loss: 1.7493 - val_acc: 0.8539 Epoch 20/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.1952 - acc: 0.9779Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 3s 523us/step - loss: 0.1946 - acc: 0.9780 - val_loss: 1.7658 - val_acc: 0.8587 Epoch 21/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1800 - acc: 0.9801Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.1774 - acc: 0.9804 - val_loss: 1.7251 - val_acc: 0.8575 Epoch 22/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1663 - acc: 0.9794Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 3s 521us/step - loss: 0.1702 - acc: 0.9789 - val_loss: 1.7634 - val_acc: 0.8599 Epoch 23/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1931 - acc: 0.9790Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 3s 523us/step - loss: 0.1951 - acc: 0.9789 - val_loss: 1.7471 - val_acc: 0.8575 Epoch 24/55 6623/6680 [============================>.] - ETA: 0s - loss: 0.2131 - acc: 0.9760Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 3s 522us/step - loss: 0.2121 - acc: 0.9760 - val_loss: 1.7049 - val_acc: 0.8539 Epoch 25/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.2002 - acc: 0.9775Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.2041 - acc: 0.9771 - val_loss: 1.7678 - val_acc: 0.8551 Epoch 26/55 6623/6680 [============================>.] - ETA: 0s - loss: 0.1690 - acc: 0.9807Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.1676 - acc: 0.9808 - val_loss: 1.7422 - val_acc: 0.8587 Epoch 27/55 6660/6680 [============================>.] - ETA: 0s - loss: 0.1964 - acc: 0.9763Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.1958 - acc: 0.9763 - val_loss: 1.7390 - val_acc: 0.8515 Epoch 28/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1691 - acc: 0.9810Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.1690 - acc: 0.9808 - val_loss: 1.6803 - val_acc: 0.8587 Epoch 29/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1747 - acc: 0.9794Epoch 00029: val_loss improved from 1.64211 to 1.63743, saving model to saved_models2/weights.best.ResNet_bs37_ep55.hdf5 6680/6680 [==============================] - 4s 528us/step - loss: 0.1723 - acc: 0.9796 - val_loss: 1.6374 - val_acc: 0.8611 Epoch 30/55 6623/6680 [============================>.] - ETA: 0s - loss: 0.1777 - acc: 0.9799Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 4s 534us/step - loss: 0.1772 - acc: 0.9798 - val_loss: 1.8212 - val_acc: 0.8539 Epoch 31/55 6623/6680 [============================>.] - ETA: 0s - loss: 0.2054 - acc: 0.9766Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.2041 - acc: 0.9766 - val_loss: 1.7776 - val_acc: 0.8563 Epoch 32/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1846 - acc: 0.9780Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.1903 - acc: 0.9775 - val_loss: 1.7635 - val_acc: 0.8623 Epoch 33/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1721 - acc: 0.9809Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.1697 - acc: 0.9811 - val_loss: 1.8039 - val_acc: 0.8587 Epoch 34/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1634 - acc: 0.9819Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.1660 - acc: 0.9819 - val_loss: 1.7976 - val_acc: 0.8563 Epoch 35/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1629 - acc: 0.9812Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 3s 524us/step - loss: 0.1618 - acc: 0.9813 - val_loss: 1.7568 - val_acc: 0.8551 Epoch 36/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1969 - acc: 0.9787Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 4s 538us/step - loss: 0.1965 - acc: 0.9789 - val_loss: 1.7379 - val_acc: 0.8599 Epoch 37/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1894 - acc: 0.9784Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 4s 540us/step - loss: 0.1891 - acc: 0.9786 - val_loss: 1.8782 - val_acc: 0.8515 Epoch 38/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1910 - acc: 0.9783Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 4s 539us/step - loss: 0.1940 - acc: 0.9781 - val_loss: 1.8125 - val_acc: 0.8599 Epoch 39/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1656 - acc: 0.9800Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.1657 - acc: 0.9801 - val_loss: 1.8360 - val_acc: 0.8515 Epoch 40/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1595 - acc: 0.9813Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 4s 535us/step - loss: 0.1573 - acc: 0.9816 - val_loss: 1.8390 - val_acc: 0.8575 Epoch 41/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1668 - acc: 0.9813Epoch 00041: val_loss did not improve 6680/6680 [==============================] - 4s 529us/step - loss: 0.1686 - acc: 0.9811 - val_loss: 1.7207 - val_acc: 0.8683 Epoch 42/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.2391 - acc: 0.9753Epoch 00042: val_loss did not improve 6680/6680 [==============================] - 3s 523us/step - loss: 0.2367 - acc: 0.9753 - val_loss: 1.7783 - val_acc: 0.8551 Epoch 43/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1705 - acc: 0.9804Epoch 00043: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.1681 - acc: 0.9807 - val_loss: 1.7614 - val_acc: 0.8647 Epoch 44/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1610 - acc: 0.9827Epoch 00044: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.1676 - acc: 0.9822 - val_loss: 1.8288 - val_acc: 0.8587 Epoch 45/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1434 - acc: 0.9851Epoch 00045: val_loss did not improve 6680/6680 [==============================] - 4s 530us/step - loss: 0.1464 - acc: 0.9849 - val_loss: 1.7054 - val_acc: 0.8587 Epoch 46/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1539 - acc: 0.9810Epoch 00046: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.1518 - acc: 0.9813 - val_loss: 1.7149 - val_acc: 0.8647 Epoch 47/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1698 - acc: 0.9818Epoch 00047: val_loss did not improve 6680/6680 [==============================] - 4s 536us/step - loss: 0.1703 - acc: 0.9816 - val_loss: 1.7665 - val_acc: 0.8623 Epoch 48/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1808 - acc: 0.9783Epoch 00048: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.1841 - acc: 0.9781 - val_loss: 1.8904 - val_acc: 0.8611 Epoch 49/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.2021 - acc: 0.9783Epoch 00049: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.2018 - acc: 0.9783 - val_loss: 1.7805 - val_acc: 0.8599 Epoch 50/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1741 - acc: 0.9800Epoch 00050: val_loss did not improve 6680/6680 [==============================] - 4s 528us/step - loss: 0.1770 - acc: 0.9796 - val_loss: 1.8164 - val_acc: 0.8575 Epoch 51/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1584 - acc: 0.9818Epoch 00051: val_loss did not improve 6680/6680 [==============================] - 4s 525us/step - loss: 0.1607 - acc: 0.9816 - val_loss: 1.7474 - val_acc: 0.8599 Epoch 52/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1666 - acc: 0.9819Epoch 00052: val_loss did not improve 6680/6680 [==============================] - 4s 531us/step - loss: 0.1696 - acc: 0.9817 - val_loss: 1.7698 - val_acc: 0.8659 Epoch 53/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1841 - acc: 0.9795Epoch 00053: val_loss did not improve 6680/6680 [==============================] - 4s 532us/step - loss: 0.1841 - acc: 0.9795 - val_loss: 1.8455 - val_acc: 0.8575 Epoch 54/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1898 - acc: 0.9803Epoch 00054: val_loss did not improve 6680/6680 [==============================] - 4s 533us/step - loss: 0.1901 - acc: 0.9802 - val_loss: 1.8572 - val_acc: 0.8539 Epoch 55/55 6586/6680 [============================>.] - ETA: 0s - loss: 0.1798 - acc: 0.9801Epoch 00055: val_loss did not improve 6680/6680 [==============================] - 3s 524us/step - loss: 0.1807 - acc: 0.9801 - val_loss: 1.7960 - val_acc: 0.8551 Batch size=40 Epoch=35 Train on 6680 samples, validate on 835 samples Epoch 1/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1636 - acc: 0.9795Epoch 00001: val_loss improved from inf to 1.65063, saving model to saved_models2/weights.best.ResNet_bs40_ep35.hdf5 6680/6680 [==============================] - 4s 527us/step - loss: 0.1633 - acc: 0.9793 - val_loss: 1.6506 - val_acc: 0.8587 Epoch 2/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1646 - acc: 0.9804Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1636 - acc: 0.9805 - val_loss: 1.7771 - val_acc: 0.8599 Epoch 3/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1544 - acc: 0.9810Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 3s 511us/step - loss: 0.1535 - acc: 0.9811 - val_loss: 1.7631 - val_acc: 0.8635 Epoch 4/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1786 - acc: 0.9798Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 3s 514us/step - loss: 0.1775 - acc: 0.9799 - val_loss: 1.8020 - val_acc: 0.8563 Epoch 5/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1872 - acc: 0.9779Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 3s 519us/step - loss: 0.1887 - acc: 0.9777 - val_loss: 1.8268 - val_acc: 0.8527 Epoch 6/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1673 - acc: 0.9825Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 3s 516us/step - loss: 0.1664 - acc: 0.9826 - val_loss: 1.8131 - val_acc: 0.8575 Epoch 7/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1683 - acc: 0.9803Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1693 - acc: 0.9802 - val_loss: 1.7990 - val_acc: 0.8575 Epoch 8/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1718 - acc: 0.9794Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1707 - acc: 0.9795 - val_loss: 1.7716 - val_acc: 0.8587 Epoch 9/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1957 - acc: 0.9783Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1965 - acc: 0.9783 - val_loss: 1.8416 - val_acc: 0.8539 Epoch 10/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1651 - acc: 0.9807Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1641 - acc: 0.9808 - val_loss: 1.8931 - val_acc: 0.8575 Epoch 11/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1755 - acc: 0.9791Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1760 - acc: 0.9790 - val_loss: 1.7870 - val_acc: 0.8647 Epoch 12/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1674 - acc: 0.9810Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1664 - acc: 0.9811 - val_loss: 1.7707 - val_acc: 0.8575 Epoch 13/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1730 - acc: 0.9812Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1744 - acc: 0.9811 - val_loss: 1.9160 - val_acc: 0.8527 Epoch 14/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1302 - acc: 0.9834Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1294 - acc: 0.9835 - val_loss: 1.7822 - val_acc: 0.8539 Epoch 15/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1979 - acc: 0.9801Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.2017 - acc: 0.9798 - val_loss: 1.7164 - val_acc: 0.8647 Epoch 16/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1896 - acc: 0.9783Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 3s 518us/step - loss: 0.1886 - acc: 0.9783 - val_loss: 1.8609 - val_acc: 0.8563 Epoch 17/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.2090 - acc: 0.9783Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 3s 511us/step - loss: 0.2102 - acc: 0.9783 - val_loss: 1.7558 - val_acc: 0.8599 Epoch 18/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1727 - acc: 0.9792Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1717 - acc: 0.9793 - val_loss: 1.7393 - val_acc: 0.8623 Epoch 19/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1721 - acc: 0.9804Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 3s 500us/step - loss: 0.1721 - acc: 0.9804 - val_loss: 1.8257 - val_acc: 0.8527 Epoch 20/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1866 - acc: 0.9801Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 3s 500us/step - loss: 0.1855 - acc: 0.9802 - val_loss: 1.8835 - val_acc: 0.8575 Epoch 21/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1883 - acc: 0.9788Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1873 - acc: 0.9787 - val_loss: 1.8583 - val_acc: 0.8527 Epoch 22/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1967 - acc: 0.9786Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1955 - acc: 0.9787 - val_loss: 1.8432 - val_acc: 0.8539 Epoch 23/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1880 - acc: 0.9801Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1869 - acc: 0.9802 - val_loss: 1.8658 - val_acc: 0.8515 Epoch 24/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1685 - acc: 0.9807Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1675 - acc: 0.9808 - val_loss: 1.8197 - val_acc: 0.8563 Epoch 25/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1522 - acc: 0.9824Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1513 - acc: 0.9825 - val_loss: 1.8538 - val_acc: 0.8527 Epoch 26/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1464 - acc: 0.9830Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1467 - acc: 0.9829 - val_loss: 1.8367 - val_acc: 0.8599 Epoch 27/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1678 - acc: 0.9816Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1717 - acc: 0.9814 - val_loss: 1.8249 - val_acc: 0.8623 Epoch 28/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1557 - acc: 0.9827Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1557 - acc: 0.9826 - val_loss: 1.9023 - val_acc: 0.8575 Epoch 29/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1675 - acc: 0.9804Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1672 - acc: 0.9804 - val_loss: 1.9087 - val_acc: 0.8455 Epoch 30/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1629 - acc: 0.9822Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 3s 509us/step - loss: 0.1643 - acc: 0.9822 - val_loss: 1.8870 - val_acc: 0.8611 Epoch 31/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1438 - acc: 0.9831Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1432 - acc: 0.9831 - val_loss: 1.9171 - val_acc: 0.8587 Epoch 32/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1633 - acc: 0.9824Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1624 - acc: 0.9825 - val_loss: 1.8538 - val_acc: 0.8587 Epoch 33/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1498 - acc: 0.9834Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1489 - acc: 0.9835 - val_loss: 1.9010 - val_acc: 0.8575 Epoch 34/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1831 - acc: 0.9812Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1820 - acc: 0.9813 - val_loss: 1.8673 - val_acc: 0.8503 Epoch 35/35 6640/6680 [============================>.] - ETA: 0s - loss: 0.1606 - acc: 0.9830Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1596 - acc: 0.9831 - val_loss: 1.9221 - val_acc: 0.8599 Batch size=40 Epoch=37 Train on 6680 samples, validate on 835 samples Epoch 1/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1830 - acc: 0.9786Epoch 00001: val_loss improved from inf to 1.69151, saving model to saved_models2/weights.best.ResNet_bs40_ep37.hdf5 6680/6680 [==============================] - 3s 514us/step - loss: 0.1843 - acc: 0.9786 - val_loss: 1.6915 - val_acc: 0.8563 Epoch 2/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1838 - acc: 0.9800Epoch 00002: val_loss improved from 1.69151 to 1.67417, saving model to saved_models2/weights.best.ResNet_bs40_ep37.hdf5 6680/6680 [==============================] - 3s 508us/step - loss: 0.1837 - acc: 0.9798 - val_loss: 1.6742 - val_acc: 0.8599 Epoch 3/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1729 - acc: 0.9797Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1718 - acc: 0.9798 - val_loss: 1.7483 - val_acc: 0.8575 Epoch 4/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1612 - acc: 0.9800Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 3s 507us/step - loss: 0.1629 - acc: 0.9798 - val_loss: 1.7521 - val_acc: 0.8563 Epoch 5/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1605 - acc: 0.9815Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1595 - acc: 0.9816 - val_loss: 1.7362 - val_acc: 0.8551 Epoch 6/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1466 - acc: 0.9815Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1467 - acc: 0.9814 - val_loss: 1.7812 - val_acc: 0.8599 Epoch 7/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1590 - acc: 0.9825Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1605 - acc: 0.9825 - val_loss: 1.7164 - val_acc: 0.8587 Epoch 8/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1742 - acc: 0.9803Epoch 00008: val_loss improved from 1.67417 to 1.67072, saving model to saved_models2/weights.best.ResNet_bs40_ep37.hdf5 6680/6680 [==============================] - 3s 513us/step - loss: 0.1744 - acc: 0.9801 - val_loss: 1.6707 - val_acc: 0.8671 Epoch 9/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1756 - acc: 0.9813Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 3s 500us/step - loss: 0.1746 - acc: 0.9814 - val_loss: 1.7530 - val_acc: 0.8635 Epoch 10/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1508 - acc: 0.9825Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 3s 507us/step - loss: 0.1499 - acc: 0.9826 - val_loss: 1.7288 - val_acc: 0.8527 Epoch 11/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1601 - acc: 0.9806Epoch 00011: val_loss improved from 1.67072 to 1.66728, saving model to saved_models2/weights.best.ResNet_bs40_ep37.hdf5 6680/6680 [==============================] - 3s 511us/step - loss: 0.1592 - acc: 0.9807 - val_loss: 1.6673 - val_acc: 0.8635 Epoch 12/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1625 - acc: 0.9821Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1615 - acc: 0.9822 - val_loss: 1.7624 - val_acc: 0.8539 Epoch 13/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1473 - acc: 0.9825Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1489 - acc: 0.9825 - val_loss: 1.7514 - val_acc: 0.8551 Epoch 14/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1642 - acc: 0.9810Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1632 - acc: 0.9811 - val_loss: 1.7240 - val_acc: 0.8527 Epoch 15/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1728 - acc: 0.9801Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1729 - acc: 0.9801 - val_loss: 1.7907 - val_acc: 0.8491 Epoch 16/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1954 - acc: 0.9794Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1942 - acc: 0.9795 - val_loss: 1.8802 - val_acc: 0.8479 Epoch 17/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1890 - acc: 0.9795Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1886 - acc: 0.9795 - val_loss: 1.9013 - val_acc: 0.8431 Epoch 18/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1599 - acc: 0.9819Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1591 - acc: 0.9819 - val_loss: 1.8346 - val_acc: 0.8479 Epoch 19/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1613 - acc: 0.9803Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 3s 511us/step - loss: 0.1603 - acc: 0.9804 - val_loss: 1.8367 - val_acc: 0.8515 Epoch 20/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1550 - acc: 0.9803Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 3s 509us/step - loss: 0.1554 - acc: 0.9802 - val_loss: 1.8498 - val_acc: 0.8407 Epoch 21/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1732 - acc: 0.9795Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1727 - acc: 0.9795 - val_loss: 1.9020 - val_acc: 0.8455 Epoch 22/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1710 - acc: 0.9810Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 3s 510us/step - loss: 0.1700 - acc: 0.9811 - val_loss: 1.8482 - val_acc: 0.8515 Epoch 23/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1599 - acc: 0.9798Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1589 - acc: 0.9799 - val_loss: 1.8940 - val_acc: 0.8443 Epoch 24/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1673 - acc: 0.9803Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1663 - acc: 0.9804 - val_loss: 1.9267 - val_acc: 0.8395 Epoch 25/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1623 - acc: 0.9804Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1637 - acc: 0.9804 - val_loss: 1.9226 - val_acc: 0.8455 Epoch 26/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1797 - acc: 0.9794Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 3s 507us/step - loss: 0.1788 - acc: 0.9793 - val_loss: 1.8297 - val_acc: 0.8479 Epoch 27/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1331 - acc: 0.9837Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1323 - acc: 0.9838 - val_loss: 1.8138 - val_acc: 0.8539 Epoch 28/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1908 - acc: 0.9798Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1897 - acc: 0.9799 - val_loss: 1.8526 - val_acc: 0.8575 Epoch 29/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1552 - acc: 0.9827Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 3s 507us/step - loss: 0.1564 - acc: 0.9825 - val_loss: 1.8998 - val_acc: 0.8575 Epoch 30/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1830 - acc: 0.9782Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1840 - acc: 0.9780 - val_loss: 1.9335 - val_acc: 0.8467 Epoch 31/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1406 - acc: 0.9839Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1397 - acc: 0.9840 - val_loss: 1.8205 - val_acc: 0.8455 Epoch 32/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1799 - acc: 0.9804Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1794 - acc: 0.9804 - val_loss: 1.9292 - val_acc: 0.8515 Epoch 33/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1591 - acc: 0.9839Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1581 - acc: 0.9840 - val_loss: 1.9346 - val_acc: 0.8491 Epoch 34/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1682 - acc: 0.9801Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1673 - acc: 0.9802 - val_loss: 1.8556 - val_acc: 0.8527 Epoch 35/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1587 - acc: 0.9822Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1609 - acc: 0.9820 - val_loss: 1.8426 - val_acc: 0.8575 Epoch 36/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1326 - acc: 0.9839Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1342 - acc: 0.9838 - val_loss: 1.8166 - val_acc: 0.8599 Epoch 37/37 6640/6680 [============================>.] - ETA: 0s - loss: 0.1583 - acc: 0.9822Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1574 - acc: 0.9823 - val_loss: 1.9435 - val_acc: 0.8515 Batch size=40 Epoch=40 Train on 6680 samples, validate on 835 samples Epoch 1/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1684 - acc: 0.9813Epoch 00001: val_loss improved from inf to 1.66598, saving model to saved_models2/weights.best.ResNet_bs40_ep40.hdf5 6680/6680 [==============================] - 3s 506us/step - loss: 0.1673 - acc: 0.9814 - val_loss: 1.6660 - val_acc: 0.8623 Epoch 2/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1761 - acc: 0.9794Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1755 - acc: 0.9793 - val_loss: 1.7194 - val_acc: 0.8671 Epoch 3/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1552 - acc: 0.9812Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 3s 517us/step - loss: 0.1558 - acc: 0.9811 - val_loss: 1.7532 - val_acc: 0.8587 Epoch 4/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1655 - acc: 0.9810Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 3s 500us/step - loss: 0.1663 - acc: 0.9810 - val_loss: 1.8292 - val_acc: 0.8539 Epoch 5/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1910 - acc: 0.9788Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1898 - acc: 0.9789 - val_loss: 1.8116 - val_acc: 0.8503 Epoch 6/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1619 - acc: 0.9822Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 3s 513us/step - loss: 0.1614 - acc: 0.9822 - val_loss: 1.8134 - val_acc: 0.8587 Epoch 7/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1455 - acc: 0.9824Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 3s 507us/step - loss: 0.1446 - acc: 0.9825 - val_loss: 1.8291 - val_acc: 0.8503 Epoch 8/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.2012 - acc: 0.9783Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 3s 509us/step - loss: 0.2000 - acc: 0.9784 - val_loss: 1.8148 - val_acc: 0.8551 Epoch 9/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1882 - acc: 0.9786Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 3s 516us/step - loss: 0.1871 - acc: 0.9787 - val_loss: 1.8112 - val_acc: 0.8587 Epoch 10/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1387 - acc: 0.9828Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1379 - acc: 0.9829 - val_loss: 1.8388 - val_acc: 0.8539 Epoch 11/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1671 - acc: 0.9809Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 3s 509us/step - loss: 0.1662 - acc: 0.9810 - val_loss: 1.7909 - val_acc: 0.8587 Epoch 12/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1885 - acc: 0.9777Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1895 - acc: 0.9777 - val_loss: 1.9115 - val_acc: 0.8491 Epoch 13/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1620 - acc: 0.9816Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 3s 498us/step - loss: 0.1614 - acc: 0.9816 - val_loss: 1.8494 - val_acc: 0.8539 Epoch 14/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1690 - acc: 0.9810Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 3s 500us/step - loss: 0.1733 - acc: 0.9807 - val_loss: 1.8676 - val_acc: 0.8515 Epoch 15/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1444 - acc: 0.9831Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1436 - acc: 0.9832 - val_loss: 1.9150 - val_acc: 0.8527 Epoch 16/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1677 - acc: 0.9816Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1691 - acc: 0.9816 - val_loss: 1.8358 - val_acc: 0.8539 Epoch 17/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1608 - acc: 0.9806Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1598 - acc: 0.9807 - val_loss: 1.7637 - val_acc: 0.8599 Epoch 18/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1540 - acc: 0.9824Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 3s 507us/step - loss: 0.1531 - acc: 0.9825 - val_loss: 1.8372 - val_acc: 0.8587 Epoch 19/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1559 - acc: 0.9816Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 3s 511us/step - loss: 0.1574 - acc: 0.9816 - val_loss: 1.8471 - val_acc: 0.8539 Epoch 20/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1776 - acc: 0.9800Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1789 - acc: 0.9799 - val_loss: 1.9047 - val_acc: 0.8503 Epoch 21/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1670 - acc: 0.9797Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1686 - acc: 0.9795 - val_loss: 1.9780 - val_acc: 0.8491 Epoch 22/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1519 - acc: 0.9833Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1521 - acc: 0.9832 - val_loss: 1.9177 - val_acc: 0.8527 Epoch 23/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1598 - acc: 0.9821Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 3s 508us/step - loss: 0.1588 - acc: 0.9822 - val_loss: 1.8832 - val_acc: 0.8539 Epoch 24/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1640 - acc: 0.9816Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 3s 507us/step - loss: 0.1658 - acc: 0.9814 - val_loss: 1.9027 - val_acc: 0.8515 Epoch 25/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1510 - acc: 0.9825Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1501 - acc: 0.9826 - val_loss: 1.8784 - val_acc: 0.8503 Epoch 26/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1614 - acc: 0.9816Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 3s 507us/step - loss: 0.1604 - acc: 0.9817 - val_loss: 1.9459 - val_acc: 0.8503 Epoch 27/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1591 - acc: 0.9824Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1581 - acc: 0.9825 - val_loss: 1.9514 - val_acc: 0.8479 Epoch 28/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1698 - acc: 0.9818Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1688 - acc: 0.9819 - val_loss: 1.9780 - val_acc: 0.8455 Epoch 29/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1810 - acc: 0.9807Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1799 - acc: 0.9808 - val_loss: 1.9489 - val_acc: 0.8455 Epoch 30/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1566 - acc: 0.9810Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1567 - acc: 0.9810 - val_loss: 1.9872 - val_acc: 0.8467 Epoch 31/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1584 - acc: 0.9827Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1574 - acc: 0.9828 - val_loss: 1.9291 - val_acc: 0.8455 Epoch 32/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1556 - acc: 0.9827Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1547 - acc: 0.9828 - val_loss: 2.0061 - val_acc: 0.8431 Epoch 33/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1867 - acc: 0.9815Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 3s 517us/step - loss: 0.1880 - acc: 0.9814 - val_loss: 1.8879 - val_acc: 0.8539 Epoch 34/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1736 - acc: 0.9819Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 3s 511us/step - loss: 0.1749 - acc: 0.9819 - val_loss: 1.9445 - val_acc: 0.8467 Epoch 35/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1611 - acc: 0.9815Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1612 - acc: 0.9814 - val_loss: 1.8904 - val_acc: 0.8551 Epoch 36/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1934 - acc: 0.9804Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1922 - acc: 0.9805 - val_loss: 1.9146 - val_acc: 0.8515 Epoch 37/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1235 - acc: 0.9857Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 3s 513us/step - loss: 0.1257 - acc: 0.9855 - val_loss: 1.8853 - val_acc: 0.8527 Epoch 38/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1511 - acc: 0.9831Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 3s 510us/step - loss: 0.1502 - acc: 0.9832 - val_loss: 1.9265 - val_acc: 0.8587 Epoch 39/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1486 - acc: 0.9849Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1478 - acc: 0.9850 - val_loss: 1.8337 - val_acc: 0.8575 Epoch 40/40 6640/6680 [============================>.] - ETA: 0s - loss: 0.1430 - acc: 0.9821Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 3s 509us/step - loss: 0.1421 - acc: 0.9822 - val_loss: 1.9012 - val_acc: 0.8611 Batch size=40 Epoch=50 Train on 6680 samples, validate on 835 samples Epoch 1/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1606 - acc: 0.9801Epoch 00001: val_loss improved from inf to 1.69763, saving model to saved_models2/weights.best.ResNet_bs40_ep50.hdf5 6680/6680 [==============================] - 3s 512us/step - loss: 0.1612 - acc: 0.9801 - val_loss: 1.6976 - val_acc: 0.8623 Epoch 2/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1546 - acc: 0.9813Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 3s 510us/step - loss: 0.1572 - acc: 0.9811 - val_loss: 1.7130 - val_acc: 0.8635 Epoch 3/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1501 - acc: 0.9822Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 3s 510us/step - loss: 0.1495 - acc: 0.9822 - val_loss: 1.8132 - val_acc: 0.8551 Epoch 4/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1669 - acc: 0.9813Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1699 - acc: 0.9808 - val_loss: 1.7761 - val_acc: 0.8611 Epoch 5/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1725 - acc: 0.9800Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 3s 497us/step - loss: 0.1715 - acc: 0.9801 - val_loss: 1.7199 - val_acc: 0.8551 Epoch 6/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1798 - acc: 0.9791Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1812 - acc: 0.9790 - val_loss: 1.8126 - val_acc: 0.8491 Epoch 7/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1775 - acc: 0.9792Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1764 - acc: 0.9793 - val_loss: 1.7861 - val_acc: 0.8587 Epoch 8/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1587 - acc: 0.9834Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1578 - acc: 0.9835 - val_loss: 1.8082 - val_acc: 0.8575 Epoch 9/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1877 - acc: 0.9789Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1866 - acc: 0.9790 - val_loss: 1.8685 - val_acc: 0.8527 Epoch 10/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1872 - acc: 0.9800Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 3s 510us/step - loss: 0.1862 - acc: 0.9799 - val_loss: 1.8273 - val_acc: 0.8539 Epoch 11/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1800 - acc: 0.9795Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 3s 508us/step - loss: 0.1790 - acc: 0.9796 - val_loss: 1.8297 - val_acc: 0.8623 Epoch 12/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1670 - acc: 0.9819Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 3s 511us/step - loss: 0.1707 - acc: 0.9816 - val_loss: 1.7558 - val_acc: 0.8587 Epoch 13/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1728 - acc: 0.9798Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 3s 513us/step - loss: 0.1738 - acc: 0.9798 - val_loss: 1.8261 - val_acc: 0.8599 Epoch 14/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.2064 - acc: 0.9783Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 3s 510us/step - loss: 0.2065 - acc: 0.9781 - val_loss: 1.8061 - val_acc: 0.8575 Epoch 15/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1855 - acc: 0.9782Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 3s 500us/step - loss: 0.1860 - acc: 0.9781 - val_loss: 1.7849 - val_acc: 0.8623 Epoch 16/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1491 - acc: 0.9830Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1482 - acc: 0.9831 - val_loss: 1.7349 - val_acc: 0.8647 Epoch 17/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1546 - acc: 0.9828Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 3s 508us/step - loss: 0.1537 - acc: 0.9829 - val_loss: 1.7691 - val_acc: 0.8623 Epoch 18/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1466 - acc: 0.9809Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 3s 510us/step - loss: 0.1460 - acc: 0.9808 - val_loss: 1.7732 - val_acc: 0.8611 Epoch 19/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1603 - acc: 0.9821Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 3s 500us/step - loss: 0.1593 - acc: 0.9822 - val_loss: 1.8276 - val_acc: 0.8599 Epoch 20/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1422 - acc: 0.9828Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1450 - acc: 0.9826 - val_loss: 1.8654 - val_acc: 0.8515 Epoch 21/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1481 - acc: 0.9827Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1496 - acc: 0.9826 - val_loss: 1.8728 - val_acc: 0.8551 Epoch 22/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1596 - acc: 0.9822Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 3s 511us/step - loss: 0.1630 - acc: 0.9819 - val_loss: 1.7234 - val_acc: 0.8599 Epoch 23/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1427 - acc: 0.9839Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 3s 507us/step - loss: 0.1419 - acc: 0.9840 - val_loss: 1.7289 - val_acc: 0.8599 Epoch 24/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1695 - acc: 0.9830Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1685 - acc: 0.9831 - val_loss: 1.7757 - val_acc: 0.8611 Epoch 25/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1656 - acc: 0.9818Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1646 - acc: 0.9819 - val_loss: 1.7789 - val_acc: 0.8539 Epoch 26/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1461 - acc: 0.9801Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 3s 513us/step - loss: 0.1473 - acc: 0.9801 - val_loss: 1.7190 - val_acc: 0.8659 Epoch 27/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1905 - acc: 0.9794Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1908 - acc: 0.9793 - val_loss: 1.7700 - val_acc: 0.8599 Epoch 28/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1735 - acc: 0.9807Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1725 - acc: 0.9808 - val_loss: 1.7738 - val_acc: 0.8611 Epoch 29/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1510 - acc: 0.9836Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1526 - acc: 0.9835 - val_loss: 1.7690 - val_acc: 0.8659 Epoch 30/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1620 - acc: 0.9819Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1610 - acc: 0.9820 - val_loss: 1.8384 - val_acc: 0.8623 Epoch 31/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1536 - acc: 0.9821Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 3s 510us/step - loss: 0.1551 - acc: 0.9820 - val_loss: 1.8316 - val_acc: 0.8575 Epoch 32/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1827 - acc: 0.9795Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 3s 507us/step - loss: 0.1816 - acc: 0.9796 - val_loss: 1.7674 - val_acc: 0.8599 Epoch 33/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1700 - acc: 0.9809Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 3s 509us/step - loss: 0.1690 - acc: 0.9810 - val_loss: 1.8325 - val_acc: 0.8551 Epoch 34/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1501 - acc: 0.9836Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 3s 509us/step - loss: 0.1492 - acc: 0.9837 - val_loss: 1.8586 - val_acc: 0.8563 Epoch 35/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1715 - acc: 0.9816Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1705 - acc: 0.9816 - val_loss: 1.8567 - val_acc: 0.8443 Epoch 36/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1801 - acc: 0.9810Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 3s 509us/step - loss: 0.1814 - acc: 0.9810 - val_loss: 1.8360 - val_acc: 0.8515 Epoch 37/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1788 - acc: 0.9798Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1778 - acc: 0.9799 - val_loss: 1.8312 - val_acc: 0.8575 Epoch 38/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1684 - acc: 0.9818Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1675 - acc: 0.9817 - val_loss: 1.8053 - val_acc: 0.8599 Epoch 39/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1914 - acc: 0.9789Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1902 - acc: 0.9790 - val_loss: 1.7880 - val_acc: 0.8539 Epoch 40/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1661 - acc: 0.9806Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1652 - acc: 0.9807 - val_loss: 1.7280 - val_acc: 0.8599 Epoch 41/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1657 - acc: 0.9812Epoch 00041: val_loss did not improve 6680/6680 [==============================] - 3s 517us/step - loss: 0.1647 - acc: 0.9813 - val_loss: 1.8008 - val_acc: 0.8599 Epoch 42/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1812 - acc: 0.9806Epoch 00042: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1801 - acc: 0.9807 - val_loss: 1.8285 - val_acc: 0.8563 Epoch 43/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1363 - acc: 0.9846Epoch 00043: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1360 - acc: 0.9846 - val_loss: 1.8122 - val_acc: 0.8659 Epoch 44/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1789 - acc: 0.9806Epoch 00044: val_loss did not improve 6680/6680 [==============================] - 3s 515us/step - loss: 0.1778 - acc: 0.9807 - val_loss: 1.8308 - val_acc: 0.8587 Epoch 45/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1495 - acc: 0.9842Epoch 00045: val_loss did not improve 6680/6680 [==============================] - 3s 508us/step - loss: 0.1486 - acc: 0.9843 - val_loss: 1.8587 - val_acc: 0.8587 Epoch 46/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1786 - acc: 0.9812Epoch 00046: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1792 - acc: 0.9811 - val_loss: 1.9461 - val_acc: 0.8515 Epoch 47/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1561 - acc: 0.9828Epoch 00047: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1552 - acc: 0.9829 - val_loss: 1.9783 - val_acc: 0.8527 Epoch 48/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1585 - acc: 0.9830Epoch 00048: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1576 - acc: 0.9831 - val_loss: 1.8344 - val_acc: 0.8503 Epoch 49/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1326 - acc: 0.9827Epoch 00049: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1350 - acc: 0.9825 - val_loss: 1.8573 - val_acc: 0.8611 Epoch 50/50 6640/6680 [============================>.] - ETA: 0s - loss: 0.1520 - acc: 0.9833Epoch 00050: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1530 - acc: 0.9832 - val_loss: 1.8341 - val_acc: 0.8527 Batch size=40 Epoch=55 Train on 6680 samples, validate on 835 samples Epoch 1/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1640 - acc: 0.9809Epoch 00001: val_loss improved from inf to 1.71222, saving model to saved_models2/weights.best.ResNet_bs40_ep55.hdf5 6680/6680 [==============================] - 3s 516us/step - loss: 0.1633 - acc: 0.9808 - val_loss: 1.7122 - val_acc: 0.8623 Epoch 2/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1723 - acc: 0.9800Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 3s 514us/step - loss: 0.1737 - acc: 0.9799 - val_loss: 1.7469 - val_acc: 0.8647 Epoch 3/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1649 - acc: 0.9807Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 3s 508us/step - loss: 0.1639 - acc: 0.9808 - val_loss: 1.7621 - val_acc: 0.8539 Epoch 4/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1821 - acc: 0.9792Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1835 - acc: 0.9790 - val_loss: 1.7132 - val_acc: 0.8635 Epoch 5/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1764 - acc: 0.9803Epoch 00005: val_loss improved from 1.71222 to 1.70566, saving model to saved_models2/weights.best.ResNet_bs40_ep55.hdf5 6680/6680 [==============================] - 3s 515us/step - loss: 0.1754 - acc: 0.9804 - val_loss: 1.7057 - val_acc: 0.8599 Epoch 6/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1774 - acc: 0.9795Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1763 - acc: 0.9796 - val_loss: 1.7540 - val_acc: 0.8647 Epoch 7/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1670 - acc: 0.9819Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 3s 512us/step - loss: 0.1684 - acc: 0.9819 - val_loss: 1.7696 - val_acc: 0.8647 Epoch 8/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1682 - acc: 0.9806Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1732 - acc: 0.9802 - val_loss: 1.8591 - val_acc: 0.8575 Epoch 9/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1593 - acc: 0.9816Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1584 - acc: 0.9817 - val_loss: 1.9046 - val_acc: 0.8527 Epoch 10/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1441 - acc: 0.9822Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 3s 508us/step - loss: 0.1432 - acc: 0.9823 - val_loss: 1.8333 - val_acc: 0.8575 Epoch 11/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1797 - acc: 0.9809Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1787 - acc: 0.9810 - val_loss: 1.8047 - val_acc: 0.8587 Epoch 12/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1762 - acc: 0.9797Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 3s 508us/step - loss: 0.1759 - acc: 0.9796 - val_loss: 1.8605 - val_acc: 0.8563 Epoch 13/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1490 - acc: 0.9825Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1534 - acc: 0.9822 - val_loss: 1.7914 - val_acc: 0.8599 Epoch 14/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1738 - acc: 0.9810Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 3s 509us/step - loss: 0.1731 - acc: 0.9810 - val_loss: 1.8240 - val_acc: 0.8539 Epoch 15/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.2018 - acc: 0.9803Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 3s 513us/step - loss: 0.2006 - acc: 0.9804 - val_loss: 1.7969 - val_acc: 0.8575 Epoch 16/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1498 - acc: 0.9822Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 3s 512us/step - loss: 0.1494 - acc: 0.9822 - val_loss: 1.9169 - val_acc: 0.8575 Epoch 17/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1799 - acc: 0.9804Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1788 - acc: 0.9805 - val_loss: 1.8616 - val_acc: 0.8587 Epoch 18/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1426 - acc: 0.9849Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1418 - acc: 0.9850 - val_loss: 1.8169 - val_acc: 0.8599 Epoch 19/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1505 - acc: 0.9812Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1496 - acc: 0.9813 - val_loss: 1.8052 - val_acc: 0.8563 Epoch 20/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1617 - acc: 0.9816Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 3s 510us/step - loss: 0.1607 - acc: 0.9817 - val_loss: 1.8527 - val_acc: 0.8515 Epoch 21/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1963 - acc: 0.9767Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1975 - acc: 0.9766 - val_loss: 1.8605 - val_acc: 0.8539 Epoch 22/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1747 - acc: 0.9807Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1756 - acc: 0.9807 - val_loss: 1.7852 - val_acc: 0.8575 Epoch 23/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1882 - acc: 0.9794Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1871 - acc: 0.9795 - val_loss: 1.8928 - val_acc: 0.8575 Epoch 24/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1677 - acc: 0.9815Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1667 - acc: 0.9816 - val_loss: 1.8719 - val_acc: 0.8527 Epoch 25/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1827 - acc: 0.9797Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1816 - acc: 0.9798 - val_loss: 1.8076 - val_acc: 0.8587 Epoch 26/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1738 - acc: 0.9803Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 3s 513us/step - loss: 0.1759 - acc: 0.9801 - val_loss: 1.7254 - val_acc: 0.8671 Epoch 27/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1616 - acc: 0.9816Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 3s 508us/step - loss: 0.1606 - acc: 0.9817 - val_loss: 1.8117 - val_acc: 0.8623 Epoch 28/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1581 - acc: 0.9825Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 3s 518us/step - loss: 0.1619 - acc: 0.9823 - val_loss: 1.7751 - val_acc: 0.8575 Epoch 29/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1437 - acc: 0.9810Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 3s 520us/step - loss: 0.1444 - acc: 0.9810 - val_loss: 1.8158 - val_acc: 0.8647 Epoch 30/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1560 - acc: 0.9818Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 3s 522us/step - loss: 0.1574 - acc: 0.9817 - val_loss: 1.8697 - val_acc: 0.8575 Epoch 31/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1593 - acc: 0.9830Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 3s 508us/step - loss: 0.1606 - acc: 0.9829 - val_loss: 1.8149 - val_acc: 0.8551 Epoch 32/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1748 - acc: 0.9806Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 3s 512us/step - loss: 0.1738 - acc: 0.9807 - val_loss: 1.9190 - val_acc: 0.8479 Epoch 33/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1505 - acc: 0.9833Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1496 - acc: 0.9834 - val_loss: 1.9040 - val_acc: 0.8455 Epoch 34/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1674 - acc: 0.9822Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1713 - acc: 0.9820 - val_loss: 1.8915 - val_acc: 0.8539 Epoch 35/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1849 - acc: 0.9807Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1862 - acc: 0.9807 - val_loss: 1.8822 - val_acc: 0.8575 Epoch 36/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1726 - acc: 0.9803Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1721 - acc: 0.9802 - val_loss: 1.8147 - val_acc: 0.8563 Epoch 37/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1403 - acc: 0.9846Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1395 - acc: 0.9847 - val_loss: 1.9024 - val_acc: 0.8563 Epoch 38/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1270 - acc: 0.9848Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 3s 507us/step - loss: 0.1263 - acc: 0.9849 - val_loss: 1.8828 - val_acc: 0.8587 Epoch 39/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1783 - acc: 0.9807Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1797 - acc: 0.9807 - val_loss: 1.9087 - val_acc: 0.8587 Epoch 40/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1656 - acc: 0.9809Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 3s 513us/step - loss: 0.1677 - acc: 0.9805 - val_loss: 1.9354 - val_acc: 0.8551 Epoch 41/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1576 - acc: 0.9818Epoch 00041: val_loss did not improve 6680/6680 [==============================] - 3s 507us/step - loss: 0.1580 - acc: 0.9816 - val_loss: 1.8835 - val_acc: 0.8551 Epoch 42/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1627 - acc: 0.9819Epoch 00042: val_loss did not improve 6680/6680 [==============================] - 3s 510us/step - loss: 0.1621 - acc: 0.9819 - val_loss: 1.8344 - val_acc: 0.8563 Epoch 43/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1613 - acc: 0.9846Epoch 00043: val_loss did not improve 6680/6680 [==============================] - 3s 516us/step - loss: 0.1608 - acc: 0.9846 - val_loss: 1.9193 - val_acc: 0.8527 Epoch 44/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1647 - acc: 0.9807Epoch 00044: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1662 - acc: 0.9807 - val_loss: 1.8708 - val_acc: 0.8539 Epoch 45/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1485 - acc: 0.9842Epoch 00045: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1501 - acc: 0.9841 - val_loss: 1.9063 - val_acc: 0.8563 Epoch 46/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1396 - acc: 0.9839Epoch 00046: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1388 - acc: 0.9840 - val_loss: 1.9169 - val_acc: 0.8539 Epoch 47/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1700 - acc: 0.9812Epoch 00047: val_loss did not improve 6680/6680 [==============================] - 3s 513us/step - loss: 0.1692 - acc: 0.9811 - val_loss: 1.8624 - val_acc: 0.8539 Epoch 48/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1420 - acc: 0.9846Epoch 00048: val_loss did not improve 6680/6680 [==============================] - 3s 507us/step - loss: 0.1412 - acc: 0.9847 - val_loss: 1.9382 - val_acc: 0.8515 Epoch 49/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1476 - acc: 0.9836Epoch 00049: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1467 - acc: 0.9837 - val_loss: 1.8462 - val_acc: 0.8575 Epoch 50/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1336 - acc: 0.9843Epoch 00050: val_loss did not improve 6680/6680 [==============================] - 3s 507us/step - loss: 0.1328 - acc: 0.9844 - val_loss: 1.9065 - val_acc: 0.8539 Epoch 51/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1482 - acc: 0.9837Epoch 00051: val_loss did not improve 6680/6680 [==============================] - 3s 508us/step - loss: 0.1508 - acc: 0.9835 - val_loss: 1.9052 - val_acc: 0.8527 Epoch 52/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1353 - acc: 0.9860Epoch 00052: val_loss did not improve 6680/6680 [==============================] - 3s 508us/step - loss: 0.1355 - acc: 0.9859 - val_loss: 2.0075 - val_acc: 0.8503 Epoch 53/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1586 - acc: 0.9840Epoch 00053: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1601 - acc: 0.9840 - val_loss: 1.9596 - val_acc: 0.8563 Epoch 54/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1805 - acc: 0.9807Epoch 00054: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1796 - acc: 0.9807 - val_loss: 2.0263 - val_acc: 0.8479 Epoch 55/55 6640/6680 [============================>.] - ETA: 0s - loss: 0.1821 - acc: 0.9812Epoch 00055: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1834 - acc: 0.9811 - val_loss: 2.0401 - val_acc: 0.8491 Batch size=41 Epoch=35 Train on 6680 samples, validate on 835 samples Epoch 1/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1645 - acc: 0.9814Epoch 00001: val_loss improved from inf to 1.64528, saving model to saved_models2/weights.best.ResNet_bs41_ep35.hdf5 6680/6680 [==============================] - 3s 507us/step - loss: 0.1664 - acc: 0.9814 - val_loss: 1.6453 - val_acc: 0.8599 Epoch 2/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1801 - acc: 0.9812Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1789 - acc: 0.9814 - val_loss: 1.7093 - val_acc: 0.8635 Epoch 3/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1812 - acc: 0.9791Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1780 - acc: 0.9795 - val_loss: 1.7455 - val_acc: 0.8623 Epoch 4/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1803 - acc: 0.9796Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1789 - acc: 0.9798 - val_loss: 1.7884 - val_acc: 0.8587 Epoch 5/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1830 - acc: 0.9802Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1849 - acc: 0.9801 - val_loss: 1.8356 - val_acc: 0.8563 Epoch 6/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1581 - acc: 0.9832Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 3s 508us/step - loss: 0.1578 - acc: 0.9834 - val_loss: 1.7747 - val_acc: 0.8611 Epoch 7/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1819 - acc: 0.9791Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 3s 500us/step - loss: 0.1820 - acc: 0.9790 - val_loss: 1.7909 - val_acc: 0.8515 Epoch 8/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1865 - acc: 0.9790Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 3s 507us/step - loss: 0.1832 - acc: 0.9793 - val_loss: 1.8131 - val_acc: 0.8623 Epoch 9/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1533 - acc: 0.9841Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 3s 496us/step - loss: 0.1559 - acc: 0.9840 - val_loss: 1.8358 - val_acc: 0.8575 Epoch 10/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1407 - acc: 0.9831Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 3s 513us/step - loss: 0.1407 - acc: 0.9831 - val_loss: 1.7999 - val_acc: 0.8515 Epoch 11/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1954 - acc: 0.9790Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 3s 508us/step - loss: 0.1920 - acc: 0.9793 - val_loss: 1.8035 - val_acc: 0.8539 Epoch 12/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1473 - acc: 0.9835Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 3s 510us/step - loss: 0.1519 - acc: 0.9834 - val_loss: 1.8544 - val_acc: 0.8551 Epoch 13/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1947 - acc: 0.9785Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1914 - acc: 0.9787 - val_loss: 1.8587 - val_acc: 0.8527 Epoch 14/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1543 - acc: 0.9814Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1515 - acc: 0.9817 - val_loss: 1.9267 - val_acc: 0.8527 Epoch 15/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1498 - acc: 0.9837Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1522 - acc: 0.9835 - val_loss: 1.8691 - val_acc: 0.8599 Epoch 16/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1604 - acc: 0.9803Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 3s 497us/step - loss: 0.1624 - acc: 0.9804 - val_loss: 1.8703 - val_acc: 0.8539 Epoch 17/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1499 - acc: 0.9849Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1552 - acc: 0.9846 - val_loss: 1.8261 - val_acc: 0.8563 Epoch 18/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1694 - acc: 0.9799Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 3s 507us/step - loss: 0.1755 - acc: 0.9796 - val_loss: 1.7643 - val_acc: 0.8515 Epoch 19/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1915 - acc: 0.9802Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1895 - acc: 0.9802 - val_loss: 1.7363 - val_acc: 0.8575 Epoch 20/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1561 - acc: 0.9828Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 3s 496us/step - loss: 0.1541 - acc: 0.9829 - val_loss: 1.7357 - val_acc: 0.8623 Epoch 21/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1684 - acc: 0.9799Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1670 - acc: 0.9799 - val_loss: 1.7545 - val_acc: 0.8599 Epoch 22/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1626 - acc: 0.9808Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1597 - acc: 0.9811 - val_loss: 1.7914 - val_acc: 0.8623 Epoch 23/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1514 - acc: 0.9819Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 3s 489us/step - loss: 0.1487 - acc: 0.9822 - val_loss: 1.7705 - val_acc: 0.8623 Epoch 24/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1458 - acc: 0.9840Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 3s 497us/step - loss: 0.1458 - acc: 0.9841 - val_loss: 1.7224 - val_acc: 0.8599 Epoch 25/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1633 - acc: 0.9823Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 3s 500us/step - loss: 0.1653 - acc: 0.9822 - val_loss: 1.7762 - val_acc: 0.8611 Epoch 26/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1511 - acc: 0.9831Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 3s 498us/step - loss: 0.1484 - acc: 0.9834 - val_loss: 1.7147 - val_acc: 0.8587 Epoch 27/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1551 - acc: 0.9829Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1574 - acc: 0.9828 - val_loss: 1.7932 - val_acc: 0.8623 Epoch 28/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1412 - acc: 0.9848Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1428 - acc: 0.9846 - val_loss: 1.7499 - val_acc: 0.8659 Epoch 29/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1652 - acc: 0.9828Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1723 - acc: 0.9822 - val_loss: 1.9082 - val_acc: 0.8563 Epoch 30/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1778 - acc: 0.9800Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 3s 496us/step - loss: 0.1746 - acc: 0.9804 - val_loss: 1.7394 - val_acc: 0.8623 Epoch 31/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1662 - acc: 0.9820Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 3s 516us/step - loss: 0.1657 - acc: 0.9822 - val_loss: 1.8042 - val_acc: 0.8599 Epoch 32/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1527 - acc: 0.9831Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 3s 514us/step - loss: 0.1550 - acc: 0.9829 - val_loss: 1.8729 - val_acc: 0.8539 Epoch 33/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1476 - acc: 0.9848Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 3s 511us/step - loss: 0.1493 - acc: 0.9844 - val_loss: 1.8252 - val_acc: 0.8515 Epoch 34/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1421 - acc: 0.9837Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 3s 510us/step - loss: 0.1403 - acc: 0.9837 - val_loss: 1.8979 - val_acc: 0.8503 Epoch 35/35 6560/6680 [============================>.] - ETA: 0s - loss: 0.1868 - acc: 0.9806Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1860 - acc: 0.9807 - val_loss: 1.8104 - val_acc: 0.8611 Batch size=41 Epoch=37 Train on 6680 samples, validate on 835 samples Epoch 1/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1662 - acc: 0.9817Epoch 00001: val_loss improved from inf to 1.77754, saving model to saved_models2/weights.best.ResNet_bs41_ep37.hdf5 6680/6680 [==============================] - 3s 509us/step - loss: 0.1657 - acc: 0.9817 - val_loss: 1.7775 - val_acc: 0.8527 Epoch 2/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1551 - acc: 0.9819Epoch 00002: val_loss improved from 1.77754 to 1.72016, saving model to saved_models2/weights.best.ResNet_bs41_ep37.hdf5 6680/6680 [==============================] - 3s 519us/step - loss: 0.1536 - acc: 0.9819 - val_loss: 1.7202 - val_acc: 0.8647 Epoch 3/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1794 - acc: 0.9790Epoch 00003: val_loss improved from 1.72016 to 1.65747, saving model to saved_models2/weights.best.ResNet_bs41_ep37.hdf5 6680/6680 [==============================] - 3s 514us/step - loss: 0.1810 - acc: 0.9790 - val_loss: 1.6575 - val_acc: 0.8635 Epoch 4/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1516 - acc: 0.9828Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1513 - acc: 0.9829 - val_loss: 1.7214 - val_acc: 0.8587 Epoch 5/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1833 - acc: 0.9791Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1833 - acc: 0.9792 - val_loss: 1.6589 - val_acc: 0.8587 Epoch 6/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1830 - acc: 0.9819Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 3s 500us/step - loss: 0.1817 - acc: 0.9817 - val_loss: 1.7618 - val_acc: 0.8551 Epoch 7/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1814 - acc: 0.9797Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1782 - acc: 0.9801 - val_loss: 1.8323 - val_acc: 0.8491 Epoch 8/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1694 - acc: 0.9809Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 3s 498us/step - loss: 0.1687 - acc: 0.9811 - val_loss: 1.7134 - val_acc: 0.8623 Epoch 9/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1508 - acc: 0.9820Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 3s 498us/step - loss: 0.1506 - acc: 0.9822 - val_loss: 1.8212 - val_acc: 0.8491 Epoch 10/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1729 - acc: 0.9806Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1698 - acc: 0.9810 - val_loss: 1.8018 - val_acc: 0.8551 Epoch 11/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1687 - acc: 0.9811Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 3s 515us/step - loss: 0.1687 - acc: 0.9811 - val_loss: 1.8171 - val_acc: 0.8503 Epoch 12/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1592 - acc: 0.9823Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 3s 498us/step - loss: 0.1636 - acc: 0.9819 - val_loss: 1.8267 - val_acc: 0.8527 Epoch 13/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1190 - acc: 0.9851Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1206 - acc: 0.9850 - val_loss: 1.7853 - val_acc: 0.8599 Epoch 14/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1830 - acc: 0.9796Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1822 - acc: 0.9796 - val_loss: 1.8008 - val_acc: 0.8539 Epoch 15/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1588 - acc: 0.9823Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1601 - acc: 0.9820 - val_loss: 1.7627 - val_acc: 0.8563 Epoch 16/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1543 - acc: 0.9832Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1546 - acc: 0.9832 - val_loss: 1.8442 - val_acc: 0.8611 Epoch 17/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1538 - acc: 0.9825Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 3s 496us/step - loss: 0.1536 - acc: 0.9825 - val_loss: 1.8406 - val_acc: 0.8491 Epoch 18/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.2001 - acc: 0.9788Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.2015 - acc: 0.9787 - val_loss: 1.8778 - val_acc: 0.8527 Epoch 19/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1685 - acc: 0.9817Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 3s 509us/step - loss: 0.1693 - acc: 0.9816 - val_loss: 1.8517 - val_acc: 0.8515 Epoch 20/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1595 - acc: 0.9829Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 3s 512us/step - loss: 0.1579 - acc: 0.9831 - val_loss: 1.8728 - val_acc: 0.8503 Epoch 21/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1771 - acc: 0.9811Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 3s 498us/step - loss: 0.1794 - acc: 0.9810 - val_loss: 1.9372 - val_acc: 0.8515 Epoch 22/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1391 - acc: 0.9848Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 3s 513us/step - loss: 0.1419 - acc: 0.9844 - val_loss: 1.8713 - val_acc: 0.8431 Epoch 23/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1725 - acc: 0.9805Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1748 - acc: 0.9802 - val_loss: 1.8276 - val_acc: 0.8551 Epoch 24/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1760 - acc: 0.9805Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1749 - acc: 0.9804 - val_loss: 1.7847 - val_acc: 0.8599 Epoch 25/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1639 - acc: 0.9816Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1646 - acc: 0.9814 - val_loss: 1.7761 - val_acc: 0.8647 Epoch 26/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1721 - acc: 0.9811Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1769 - acc: 0.9808 - val_loss: 1.7718 - val_acc: 0.8599 Epoch 27/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1759 - acc: 0.9812Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 3s 508us/step - loss: 0.1728 - acc: 0.9816 - val_loss: 1.8532 - val_acc: 0.8563 Epoch 28/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1341 - acc: 0.9819Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1341 - acc: 0.9820 - val_loss: 1.7629 - val_acc: 0.8515 Epoch 29/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1458 - acc: 0.9840Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1515 - acc: 0.9835 - val_loss: 1.8086 - val_acc: 0.8503 Epoch 30/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1849 - acc: 0.9805Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 3s 498us/step - loss: 0.1827 - acc: 0.9804 - val_loss: 1.8602 - val_acc: 0.8503 Epoch 31/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1627 - acc: 0.9829Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 3s 515us/step - loss: 0.1598 - acc: 0.9832 - val_loss: 1.7952 - val_acc: 0.8527 Epoch 32/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1669 - acc: 0.9816Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1697 - acc: 0.9814 - val_loss: 1.8364 - val_acc: 0.8527 Epoch 33/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1380 - acc: 0.9838Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 3s 508us/step - loss: 0.1403 - acc: 0.9838 - val_loss: 1.8874 - val_acc: 0.8539 Epoch 34/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1687 - acc: 0.9802Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 3s 510us/step - loss: 0.1659 - acc: 0.9802 - val_loss: 1.8252 - val_acc: 0.8527 Epoch 35/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1672 - acc: 0.9814Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1660 - acc: 0.9814 - val_loss: 1.8514 - val_acc: 0.8539 Epoch 36/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1519 - acc: 0.9835Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 3s 493us/step - loss: 0.1519 - acc: 0.9832 - val_loss: 1.8333 - val_acc: 0.8527 Epoch 37/37 6560/6680 [============================>.] - ETA: 0s - loss: 0.1724 - acc: 0.9816Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1706 - acc: 0.9817 - val_loss: 1.6981 - val_acc: 0.8587 Batch size=41 Epoch=40 Train on 6680 samples, validate on 835 samples Epoch 1/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1572 - acc: 0.9828Epoch 00001: val_loss improved from inf to 1.71220, saving model to saved_models2/weights.best.ResNet_bs41_ep40.hdf5 6680/6680 [==============================] - 3s 515us/step - loss: 0.1558 - acc: 0.9826 - val_loss: 1.7122 - val_acc: 0.8647 Epoch 2/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1703 - acc: 0.9806Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1681 - acc: 0.9808 - val_loss: 1.7373 - val_acc: 0.8587 Epoch 3/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1655 - acc: 0.9817Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1627 - acc: 0.9820 - val_loss: 1.7543 - val_acc: 0.8683 Epoch 4/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1816 - acc: 0.9799Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1784 - acc: 0.9802 - val_loss: 1.7645 - val_acc: 0.8635 Epoch 5/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1585 - acc: 0.9823Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 3s 500us/step - loss: 0.1605 - acc: 0.9823 - val_loss: 1.7530 - val_acc: 0.8635 Epoch 6/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1842 - acc: 0.9803Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1870 - acc: 0.9801 - val_loss: 1.7614 - val_acc: 0.8635 Epoch 7/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1783 - acc: 0.9806Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1775 - acc: 0.9808 - val_loss: 1.7600 - val_acc: 0.8707 Epoch 8/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1530 - acc: 0.9828Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1503 - acc: 0.9831 - val_loss: 1.7361 - val_acc: 0.8635 Epoch 9/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1895 - acc: 0.9790Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1861 - acc: 0.9793 - val_loss: 1.8424 - val_acc: 0.8539 Epoch 10/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1880 - acc: 0.9791Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1911 - acc: 0.9786 - val_loss: 1.7495 - val_acc: 0.8551 Epoch 11/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1870 - acc: 0.9805Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 3s 507us/step - loss: 0.1922 - acc: 0.9802 - val_loss: 1.7781 - val_acc: 0.8623 Epoch 12/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1577 - acc: 0.9829Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1585 - acc: 0.9828 - val_loss: 1.7773 - val_acc: 0.8575 Epoch 13/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1487 - acc: 0.9831Epoch 00013: val_loss improved from 1.71220 to 1.63771, saving model to saved_models2/weights.best.ResNet_bs41_ep40.hdf5 6680/6680 [==============================] - 3s 521us/step - loss: 0.1492 - acc: 0.9829 - val_loss: 1.6377 - val_acc: 0.8647 Epoch 14/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1672 - acc: 0.9812Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1682 - acc: 0.9811 - val_loss: 1.7317 - val_acc: 0.8599 Epoch 15/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1600 - acc: 0.9826Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1572 - acc: 0.9829 - val_loss: 1.7548 - val_acc: 0.8611 Epoch 16/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1341 - acc: 0.9845Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 3s 513us/step - loss: 0.1337 - acc: 0.9844 - val_loss: 1.8233 - val_acc: 0.8551 Epoch 17/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1643 - acc: 0.9837Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 3s 514us/step - loss: 0.1664 - acc: 0.9835 - val_loss: 1.7456 - val_acc: 0.8587 Epoch 18/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1616 - acc: 0.9829Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 3s 494us/step - loss: 0.1660 - acc: 0.9825 - val_loss: 1.7727 - val_acc: 0.8551 Epoch 19/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1520 - acc: 0.9825Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1523 - acc: 0.9823 - val_loss: 1.8884 - val_acc: 0.8539 Epoch 20/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1572 - acc: 0.9817Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 3s 508us/step - loss: 0.1545 - acc: 0.9820 - val_loss: 1.8920 - val_acc: 0.8575 Epoch 21/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1501 - acc: 0.9820Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 3s 496us/step - loss: 0.1565 - acc: 0.9816 - val_loss: 1.9248 - val_acc: 0.8527 Epoch 22/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1529 - acc: 0.9837Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 3s 495us/step - loss: 0.1525 - acc: 0.9838 - val_loss: 1.8117 - val_acc: 0.8587 Epoch 23/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1754 - acc: 0.9820Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1776 - acc: 0.9819 - val_loss: 1.8790 - val_acc: 0.8575 Epoch 24/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1705 - acc: 0.9817Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 3s 496us/step - loss: 0.1754 - acc: 0.9813 - val_loss: 1.8030 - val_acc: 0.8623 Epoch 25/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1633 - acc: 0.9814Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1673 - acc: 0.9811 - val_loss: 1.8461 - val_acc: 0.8527 Epoch 26/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1456 - acc: 0.9841Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1475 - acc: 0.9838 - val_loss: 1.8279 - val_acc: 0.8503 Epoch 27/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1495 - acc: 0.9838Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1505 - acc: 0.9838 - val_loss: 1.7921 - val_acc: 0.8551 Epoch 28/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1278 - acc: 0.9860Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1256 - acc: 0.9862 - val_loss: 1.8407 - val_acc: 0.8551 Epoch 29/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1630 - acc: 0.9834Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1610 - acc: 0.9835 - val_loss: 1.8389 - val_acc: 0.8623 Epoch 30/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1656 - acc: 0.9817Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 3s 498us/step - loss: 0.1722 - acc: 0.9813 - val_loss: 1.9057 - val_acc: 0.8515 Epoch 31/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1500 - acc: 0.9834Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1485 - acc: 0.9835 - val_loss: 1.7574 - val_acc: 0.8683 Epoch 32/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1509 - acc: 0.9832Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1545 - acc: 0.9831 - val_loss: 1.7156 - val_acc: 0.8695 Epoch 33/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1448 - acc: 0.9857Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1466 - acc: 0.9855 - val_loss: 1.8730 - val_acc: 0.8551 Epoch 34/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1649 - acc: 0.9812Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 3s 515us/step - loss: 0.1641 - acc: 0.9813 - val_loss: 1.8926 - val_acc: 0.8491 Epoch 35/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1671 - acc: 0.9817Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 3s 494us/step - loss: 0.1673 - acc: 0.9816 - val_loss: 1.9162 - val_acc: 0.8503 Epoch 36/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1737 - acc: 0.9803Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 3s 508us/step - loss: 0.1781 - acc: 0.9799 - val_loss: 1.8548 - val_acc: 0.8515 Epoch 37/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1588 - acc: 0.9826Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1586 - acc: 0.9826 - val_loss: 1.8833 - val_acc: 0.8443 Epoch 38/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1326 - acc: 0.9835Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 3s 507us/step - loss: 0.1302 - acc: 0.9838 - val_loss: 1.8256 - val_acc: 0.8491 Epoch 39/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1366 - acc: 0.9841Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 3s 516us/step - loss: 0.1389 - acc: 0.9841 - val_loss: 1.8739 - val_acc: 0.8575 Epoch 40/40 6560/6680 [============================>.] - ETA: 0s - loss: 0.1383 - acc: 0.9846Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1382 - acc: 0.9846 - val_loss: 1.8535 - val_acc: 0.8587 Batch size=41 Epoch=50 Train on 6680 samples, validate on 835 samples Epoch 1/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1903 - acc: 0.9791Epoch 00001: val_loss improved from inf to 1.65370, saving model to saved_models2/weights.best.ResNet_bs41_ep50.hdf5 6680/6680 [==============================] - 3s 505us/step - loss: 0.1908 - acc: 0.9790 - val_loss: 1.6537 - val_acc: 0.8683 Epoch 2/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1607 - acc: 0.9817Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 3s 508us/step - loss: 0.1598 - acc: 0.9817 - val_loss: 1.7019 - val_acc: 0.8563 Epoch 3/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1320 - acc: 0.9848Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1328 - acc: 0.9846 - val_loss: 1.7150 - val_acc: 0.8587 Epoch 4/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1677 - acc: 0.9809Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 3s 498us/step - loss: 0.1709 - acc: 0.9807 - val_loss: 1.7345 - val_acc: 0.8539 Epoch 5/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1556 - acc: 0.9816Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1533 - acc: 0.9817 - val_loss: 1.8734 - val_acc: 0.8515 Epoch 6/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1659 - acc: 0.9809Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 3s 512us/step - loss: 0.1652 - acc: 0.9808 - val_loss: 1.7937 - val_acc: 0.8551 Epoch 7/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1950 - acc: 0.9791Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1917 - acc: 0.9793 - val_loss: 1.8296 - val_acc: 0.8587 Epoch 8/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1591 - acc: 0.9816Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 3s 510us/step - loss: 0.1562 - acc: 0.9819 - val_loss: 1.8401 - val_acc: 0.8563 Epoch 9/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1598 - acc: 0.9838Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1607 - acc: 0.9835 - val_loss: 1.7646 - val_acc: 0.8575 Epoch 10/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1690 - acc: 0.9817Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1664 - acc: 0.9819 - val_loss: 1.7884 - val_acc: 0.8563 Epoch 11/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1414 - acc: 0.9857Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1404 - acc: 0.9858 - val_loss: 1.7427 - val_acc: 0.8527 Epoch 12/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1484 - acc: 0.9825Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1475 - acc: 0.9826 - val_loss: 1.7631 - val_acc: 0.8503 Epoch 13/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1588 - acc: 0.9822Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 3s 493us/step - loss: 0.1571 - acc: 0.9823 - val_loss: 1.8621 - val_acc: 0.8527 Epoch 14/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1461 - acc: 0.9832Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 3s 497us/step - loss: 0.1509 - acc: 0.9829 - val_loss: 1.7965 - val_acc: 0.8623 Epoch 15/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1521 - acc: 0.9840Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 3s 509us/step - loss: 0.1554 - acc: 0.9838 - val_loss: 1.7679 - val_acc: 0.8551 Epoch 16/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1715 - acc: 0.9803Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1689 - acc: 0.9805 - val_loss: 1.8780 - val_acc: 0.8551 Epoch 17/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1612 - acc: 0.9832Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 3s 498us/step - loss: 0.1583 - acc: 0.9835 - val_loss: 1.8182 - val_acc: 0.8707 Epoch 18/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1565 - acc: 0.9828Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1574 - acc: 0.9826 - val_loss: 1.8425 - val_acc: 0.8599 Epoch 19/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1692 - acc: 0.9823Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1682 - acc: 0.9823 - val_loss: 1.8540 - val_acc: 0.8551 Epoch 20/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1326 - acc: 0.9848Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1369 - acc: 0.9841 - val_loss: 1.8213 - val_acc: 0.8611 Epoch 21/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1646 - acc: 0.9823Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 3s 516us/step - loss: 0.1620 - acc: 0.9825 - val_loss: 1.8165 - val_acc: 0.8647 Epoch 22/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1699 - acc: 0.9811Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 3s 509us/step - loss: 0.1683 - acc: 0.9813 - val_loss: 1.8391 - val_acc: 0.8659 Epoch 23/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1571 - acc: 0.9834Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1543 - acc: 0.9837 - val_loss: 1.8831 - val_acc: 0.8623 Epoch 24/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1631 - acc: 0.9826Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 3s 507us/step - loss: 0.1602 - acc: 0.9829 - val_loss: 1.8334 - val_acc: 0.8635 Epoch 25/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1573 - acc: 0.9848Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 3s 498us/step - loss: 0.1579 - acc: 0.9847 - val_loss: 1.8277 - val_acc: 0.8611 Epoch 26/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1679 - acc: 0.9814Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 3s 500us/step - loss: 0.1668 - acc: 0.9813 - val_loss: 1.7557 - val_acc: 0.8647 Epoch 27/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1713 - acc: 0.9831Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1708 - acc: 0.9831 - val_loss: 1.7893 - val_acc: 0.8635 Epoch 28/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1532 - acc: 0.9837Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1543 - acc: 0.9834 - val_loss: 1.9391 - val_acc: 0.8551 Epoch 29/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1858 - acc: 0.9800Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 3s 517us/step - loss: 0.1867 - acc: 0.9801 - val_loss: 1.8824 - val_acc: 0.8599 Epoch 30/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1617 - acc: 0.9822Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 3s 500us/step - loss: 0.1588 - acc: 0.9825 - val_loss: 1.8551 - val_acc: 0.8587 Epoch 31/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1503 - acc: 0.9837Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 3s 496us/step - loss: 0.1476 - acc: 0.9840 - val_loss: 1.8314 - val_acc: 0.8587 Epoch 32/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1514 - acc: 0.9828Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1535 - acc: 0.9828 - val_loss: 1.8959 - val_acc: 0.8611 Epoch 33/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1462 - acc: 0.9834Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1510 - acc: 0.9831 - val_loss: 1.8666 - val_acc: 0.8659 Epoch 34/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1542 - acc: 0.9817Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 3s 508us/step - loss: 0.1555 - acc: 0.9814 - val_loss: 1.8398 - val_acc: 0.8635 Epoch 35/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1530 - acc: 0.9845Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 3s 498us/step - loss: 0.1530 - acc: 0.9844 - val_loss: 1.8540 - val_acc: 0.8551 Epoch 36/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1265 - acc: 0.9838Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1293 - acc: 0.9837 - val_loss: 1.8117 - val_acc: 0.8623 Epoch 37/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1372 - acc: 0.9841Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1348 - acc: 0.9844 - val_loss: 1.8175 - val_acc: 0.8563 Epoch 38/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1404 - acc: 0.9855Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 3s 498us/step - loss: 0.1406 - acc: 0.9853 - val_loss: 1.7926 - val_acc: 0.8623 Epoch 39/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1431 - acc: 0.9852Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1430 - acc: 0.9853 - val_loss: 1.7772 - val_acc: 0.8575 Epoch 40/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1618 - acc: 0.9829Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1613 - acc: 0.9831 - val_loss: 1.7832 - val_acc: 0.8575 Epoch 41/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1686 - acc: 0.9811Epoch 00041: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1724 - acc: 0.9810 - val_loss: 1.7558 - val_acc: 0.8599 Epoch 42/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1409 - acc: 0.9832Epoch 00042: val_loss did not improve 6680/6680 [==============================] - 3s 508us/step - loss: 0.1409 - acc: 0.9834 - val_loss: 1.9158 - val_acc: 0.8515 Epoch 43/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1551 - acc: 0.9819Epoch 00043: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1591 - acc: 0.9817 - val_loss: 1.8460 - val_acc: 0.8503 Epoch 44/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1756 - acc: 0.9831Epoch 00044: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1756 - acc: 0.9831 - val_loss: 1.7546 - val_acc: 0.8587 Epoch 45/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1336 - acc: 0.9860Epoch 00045: val_loss did not improve 6680/6680 [==============================] - 3s 498us/step - loss: 0.1312 - acc: 0.9862 - val_loss: 1.8531 - val_acc: 0.8551 Epoch 46/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1741 - acc: 0.9817Epoch 00046: val_loss did not improve 6680/6680 [==============================] - 3s 512us/step - loss: 0.1728 - acc: 0.9819 - val_loss: 1.8555 - val_acc: 0.8623 Epoch 47/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1716 - acc: 0.9828Epoch 00047: val_loss did not improve 6680/6680 [==============================] - 3s 500us/step - loss: 0.1720 - acc: 0.9826 - val_loss: 1.8122 - val_acc: 0.8563 Epoch 48/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1510 - acc: 0.9840Epoch 00048: val_loss did not improve 6680/6680 [==============================] - 3s 500us/step - loss: 0.1483 - acc: 0.9843 - val_loss: 1.7970 - val_acc: 0.8551 Epoch 49/50 6560/6680 [============================>.] - ETA: 0s - loss: 0.1635 - acc: 0.9835Epoch 00049: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1629 - acc: 0.9835 - val_loss: 1.8363 - val_acc: 0.8623 Epoch 50/50 6642/6680 [============================>.] - ETA: 0s - loss: 0.1562 - acc: 0.9842Epoch 00050: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1562 - acc: 0.9841 - val_loss: 1.7905 - val_acc: 0.8551 Batch size=41 Epoch=55 Train on 6680 samples, validate on 835 samples Epoch 1/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1486 - acc: 0.9829Epoch 00001: val_loss improved from inf to 1.72033, saving model to saved_models2/weights.best.ResNet_bs41_ep55.hdf5 6680/6680 [==============================] - 3s 510us/step - loss: 0.1582 - acc: 0.9822 - val_loss: 1.7203 - val_acc: 0.8575 Epoch 2/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1579 - acc: 0.9828Epoch 00002: val_loss did not improve 6680/6680 [==============================] - 3s 500us/step - loss: 0.1571 - acc: 0.9829 - val_loss: 1.7714 - val_acc: 0.8563 Epoch 3/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1630 - acc: 0.9816Epoch 00003: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1624 - acc: 0.9817 - val_loss: 1.8058 - val_acc: 0.8575 Epoch 4/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1551 - acc: 0.9831Epoch 00004: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1557 - acc: 0.9831 - val_loss: 1.8205 - val_acc: 0.8563 Epoch 5/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1768 - acc: 0.9808Epoch 00005: val_loss did not improve 6680/6680 [==============================] - 3s 509us/step - loss: 0.1738 - acc: 0.9810 - val_loss: 1.8278 - val_acc: 0.8539 Epoch 6/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1656 - acc: 0.9820Epoch 00006: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1650 - acc: 0.9822 - val_loss: 1.8254 - val_acc: 0.8575 Epoch 7/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1763 - acc: 0.9803Epoch 00007: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1754 - acc: 0.9804 - val_loss: 1.8145 - val_acc: 0.8515 Epoch 8/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1874 - acc: 0.9800Epoch 00008: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1864 - acc: 0.9802 - val_loss: 1.8545 - val_acc: 0.8563 Epoch 9/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1519 - acc: 0.9814Epoch 00009: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1544 - acc: 0.9813 - val_loss: 1.9141 - val_acc: 0.8503 Epoch 10/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1330 - acc: 0.9846Epoch 00010: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1359 - acc: 0.9844 - val_loss: 1.8302 - val_acc: 0.8563 Epoch 11/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1860 - acc: 0.9793Epoch 00011: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1854 - acc: 0.9793 - val_loss: 1.8056 - val_acc: 0.8515 Epoch 12/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1448 - acc: 0.9846Epoch 00012: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1426 - acc: 0.9847 - val_loss: 1.8080 - val_acc: 0.8563 Epoch 13/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1398 - acc: 0.9820Epoch 00013: val_loss did not improve 6680/6680 [==============================] - 3s 504us/step - loss: 0.1376 - acc: 0.9820 - val_loss: 1.8067 - val_acc: 0.8599 Epoch 14/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1483 - acc: 0.9848Epoch 00014: val_loss did not improve 6680/6680 [==============================] - 3s 496us/step - loss: 0.1470 - acc: 0.9846 - val_loss: 1.8980 - val_acc: 0.8407 Epoch 15/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1551 - acc: 0.9826Epoch 00015: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1555 - acc: 0.9826 - val_loss: 1.8853 - val_acc: 0.8503 Epoch 16/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1601 - acc: 0.9831Epoch 00016: val_loss did not improve 6680/6680 [==============================] - 3s 501us/step - loss: 0.1632 - acc: 0.9826 - val_loss: 1.8587 - val_acc: 0.8551 Epoch 17/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1484 - acc: 0.9841Epoch 00017: val_loss did not improve 6680/6680 [==============================] - 3s 497us/step - loss: 0.1459 - acc: 0.9844 - val_loss: 1.8354 - val_acc: 0.8551 Epoch 18/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1610 - acc: 0.9822Epoch 00018: val_loss did not improve 6680/6680 [==============================] - 3s 498us/step - loss: 0.1618 - acc: 0.9822 - val_loss: 1.8707 - val_acc: 0.8599 Epoch 19/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1774 - acc: 0.9811Epoch 00019: val_loss did not improve 6680/6680 [==============================] - 3s 496us/step - loss: 0.1770 - acc: 0.9811 - val_loss: 1.9307 - val_acc: 0.8527 Epoch 20/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1452 - acc: 0.9831Epoch 00020: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1426 - acc: 0.9834 - val_loss: 1.8316 - val_acc: 0.8563 Epoch 21/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1484 - acc: 0.9826Epoch 00021: val_loss did not improve 6680/6680 [==============================] - 3s 495us/step - loss: 0.1512 - acc: 0.9825 - val_loss: 1.9176 - val_acc: 0.8503 Epoch 22/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1700 - acc: 0.9797Epoch 00022: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1708 - acc: 0.9798 - val_loss: 1.8640 - val_acc: 0.8587 Epoch 23/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1740 - acc: 0.9812Epoch 00023: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1709 - acc: 0.9816 - val_loss: 1.8368 - val_acc: 0.8575 Epoch 24/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1580 - acc: 0.9841Epoch 00024: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1552 - acc: 0.9844 - val_loss: 1.8444 - val_acc: 0.8551 Epoch 25/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1554 - acc: 0.9828Epoch 00025: val_loss did not improve 6680/6680 [==============================] - 3s 507us/step - loss: 0.1551 - acc: 0.9829 - val_loss: 1.8432 - val_acc: 0.8539 Epoch 26/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1361 - acc: 0.9858Epoch 00026: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1418 - acc: 0.9855 - val_loss: 1.8263 - val_acc: 0.8575 Epoch 27/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1609 - acc: 0.9823Epoch 00027: val_loss did not improve 6680/6680 [==============================] - 3s 510us/step - loss: 0.1616 - acc: 0.9823 - val_loss: 1.8124 - val_acc: 0.8539 Epoch 28/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1587 - acc: 0.9828Epoch 00028: val_loss did not improve 6680/6680 [==============================] - 3s 511us/step - loss: 0.1605 - acc: 0.9826 - val_loss: 1.8222 - val_acc: 0.8575 Epoch 29/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1758 - acc: 0.9814Epoch 00029: val_loss did not improve 6680/6680 [==============================] - 3s 511us/step - loss: 0.1777 - acc: 0.9810 - val_loss: 1.7615 - val_acc: 0.8551 Epoch 30/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1695 - acc: 0.9829Epoch 00030: val_loss did not improve 6680/6680 [==============================] - 3s 496us/step - loss: 0.1709 - acc: 0.9825 - val_loss: 1.8649 - val_acc: 0.8575 Epoch 31/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1621 - acc: 0.9828Epoch 00031: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1634 - acc: 0.9828 - val_loss: 1.8051 - val_acc: 0.8635 Epoch 32/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1542 - acc: 0.9823Epoch 00032: val_loss did not improve 6680/6680 [==============================] - 3s 498us/step - loss: 0.1539 - acc: 0.9825 - val_loss: 1.9162 - val_acc: 0.8527 Epoch 33/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1622 - acc: 0.9846Epoch 00033: val_loss did not improve 6680/6680 [==============================] - 3s 508us/step - loss: 0.1607 - acc: 0.9847 - val_loss: 1.8583 - val_acc: 0.8551 Epoch 34/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1976 - acc: 0.9787Epoch 00034: val_loss did not improve 6680/6680 [==============================] - 3s 495us/step - loss: 0.1993 - acc: 0.9784 - val_loss: 1.8949 - val_acc: 0.8563 Epoch 35/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1898 - acc: 0.9797Epoch 00035: val_loss did not improve 6680/6680 [==============================] - 3s 505us/step - loss: 0.1899 - acc: 0.9798 - val_loss: 1.7869 - val_acc: 0.8575 Epoch 36/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1701 - acc: 0.9823Epoch 00036: val_loss did not improve 6680/6680 [==============================] - 3s 494us/step - loss: 0.1732 - acc: 0.9822 - val_loss: 1.7569 - val_acc: 0.8635 Epoch 37/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1618 - acc: 0.9834Epoch 00037: val_loss did not improve 6680/6680 [==============================] - 3s 496us/step - loss: 0.1609 - acc: 0.9832 - val_loss: 1.7923 - val_acc: 0.8587 Epoch 38/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1696 - acc: 0.9812Epoch 00038: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1698 - acc: 0.9811 - val_loss: 1.9269 - val_acc: 0.8515 Epoch 39/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1703 - acc: 0.9802Epoch 00039: val_loss did not improve 6680/6680 [==============================] - 3s 506us/step - loss: 0.1676 - acc: 0.9802 - val_loss: 1.8772 - val_acc: 0.8611 Epoch 40/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1548 - acc: 0.9828Epoch 00040: val_loss did not improve 6680/6680 [==============================] - 3s 497us/step - loss: 0.1567 - acc: 0.9826 - val_loss: 1.8142 - val_acc: 0.8551 Epoch 41/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1666 - acc: 0.9825Epoch 00041: val_loss did not improve 6680/6680 [==============================] - 3s 498us/step - loss: 0.1661 - acc: 0.9826 - val_loss: 1.8720 - val_acc: 0.8551 Epoch 42/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1536 - acc: 0.9841Epoch 00042: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1541 - acc: 0.9841 - val_loss: 1.8437 - val_acc: 0.8587 Epoch 43/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1496 - acc: 0.9849Epoch 00043: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1500 - acc: 0.9847 - val_loss: 1.8672 - val_acc: 0.8527 Epoch 44/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1551 - acc: 0.9835Epoch 00044: val_loss did not improve 6680/6680 [==============================] - 3s 500us/step - loss: 0.1542 - acc: 0.9834 - val_loss: 1.9599 - val_acc: 0.8539 Epoch 45/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1636 - acc: 0.9832Epoch 00045: val_loss did not improve 6680/6680 [==============================] - 3s 500us/step - loss: 0.1631 - acc: 0.9834 - val_loss: 1.9295 - val_acc: 0.8551 Epoch 46/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1820 - acc: 0.9808Epoch 00046: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1811 - acc: 0.9810 - val_loss: 1.9263 - val_acc: 0.8515 Epoch 47/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1631 - acc: 0.9828Epoch 00047: val_loss did not improve 6680/6680 [==============================] - 3s 497us/step - loss: 0.1603 - acc: 0.9829 - val_loss: 1.8972 - val_acc: 0.8503 Epoch 48/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1577 - acc: 0.9831Epoch 00048: val_loss did not improve 6680/6680 [==============================] - 3s 500us/step - loss: 0.1562 - acc: 0.9832 - val_loss: 1.8337 - val_acc: 0.8659 Epoch 49/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1695 - acc: 0.9822Epoch 00049: val_loss did not improve 6680/6680 [==============================] - 3s 497us/step - loss: 0.1694 - acc: 0.9822 - val_loss: 1.9263 - val_acc: 0.8575 Epoch 50/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1698 - acc: 0.9832Epoch 00050: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1669 - acc: 0.9835 - val_loss: 1.9187 - val_acc: 0.8515 Epoch 51/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1561 - acc: 0.9831Epoch 00051: val_loss did not improve 6680/6680 [==============================] - 3s 502us/step - loss: 0.1594 - acc: 0.9829 - val_loss: 1.9099 - val_acc: 0.8587 Epoch 52/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1482 - acc: 0.9837Epoch 00052: val_loss did not improve 6680/6680 [==============================] - 3s 499us/step - loss: 0.1470 - acc: 0.9837 - val_loss: 2.0299 - val_acc: 0.8479 Epoch 53/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1448 - acc: 0.9837Epoch 00053: val_loss did not improve 6680/6680 [==============================] - 3s 496us/step - loss: 0.1436 - acc: 0.9838 - val_loss: 1.9781 - val_acc: 0.8491 Epoch 54/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1288 - acc: 0.9852Epoch 00054: val_loss did not improve 6680/6680 [==============================] - 3s 503us/step - loss: 0.1274 - acc: 0.9853 - val_loss: 1.8549 - val_acc: 0.8599 Epoch 55/55 6560/6680 [============================>.] - ETA: 0s - loss: 0.1470 - acc: 0.9832Epoch 00055: val_loss did not improve 6680/6680 [==============================] - 3s 496us/step - loss: 0.1444 - acc: 0.9835 - val_loss: 1.7842 - val_acc: 0.8611
In [56]:
pd.DataFrame(fitingdict)
Out[56]:
| Batch_Size | Epochs | Test_Accuracy | |
|---|---|---|---|
| 0 | 35 | 35 | 85.406699 |
| 1 | 35 | 37 | 85.287081 |
| 2 | 35 | 40 | 84.928230 |
| 3 | 35 | 50 | 85.645933 |
| 4 | 35 | 55 | 85.287081 |
| 5 | 36 | 35 | 84.330144 |
| 6 | 36 | 37 | 84.928230 |
| 7 | 36 | 40 | 84.808612 |
| 8 | 36 | 50 | 85.645933 |
| 9 | 36 | 55 | 85.885167 |
| 10 | 37 | 35 | 85.765550 |
| 11 | 37 | 37 | 85.287081 |
| 12 | 37 | 40 | 85.645933 |
| 13 | 37 | 50 | 85.645933 |
| 14 | 37 | 55 | 85.406699 |
| 15 | 40 | 35 | 85.287081 |
| 16 | 40 | 37 | 85.645933 |
| 17 | 40 | 40 | 86.363636 |
| 18 | 40 | 50 | 86.124402 |
| 19 | 40 | 55 | 85.765550 |
| 20 | 41 | 35 | 85.765550 |
| 21 | 41 | 37 | 85.765550 |
| 22 | 41 | 40 | 86.602871 |
| 23 | 41 | 50 | 86.842105 |
| 24 | 41 | 55 | 86.004785 |
(IMPLEMENTATION) Load the Model with the Best Validation Loss¶
In [57]:
#take largest testaccuracy's batch size and epochs
ind=fitingdict['Test_Accuracy'].index(max(fitingdict['Test_Accuracy']))
bs=fitingdict['Batch_Size'][ind]
ep=fitingdict['Epochs'][ind]
#LOAD the model with Best validation loss
Xception_model.load_weights('saved_models2/weights.best.ResNet_bs'+str(bs)+'_ep'+str(ep)+'.hdf5')
(IMPLEMENTATION) Test the Model¶
Try out your model on the test dataset of dog images. Ensure that your test accuracy is greater than 60%.
In [30]:
### TODO: Calculate classification accuracy on the test dataset.
# get index of predicted dog breed for each image in test set
Xception_predictions = [np.argmax(Xception_model.predict(np.expand_dims(feature, axis=0))) for feature in test_Xception]
# report test accuracy
test_accuracy = 100*np.sum(np.array(Xception_predictions)==np.argmax(test_targets, axis=1))/len(Xception_predictions)
print('Test accuracy: %.4f%%' % test_accuracy)
Test accuracy: 86.8421%
In [59]:
test_Xception[0].shape
Out[59]:
(7, 7, 2048)
In [60]:
len(Xception_model.predict(np.expand_dims(test_Xception[0], axis=0))[0])
Out[60]:
133
In [61]:
len(np.array(Xception_predictions))
Out[61]:
836
In [62]:
len(np.argmax(test_targets, axis=1))
Out[62]:
836
In [63]:
dog_names[np.argmax(test_targets,axis=1)[1]]
Out[63]:
'Doberman_pinscher'
(IMPLEMENTATION) Predict Dog Breed with the Model¶
Write a function that takes an image path as input and returns the dog breed (Affenpinscher, Afghan_hound, etc) that is predicted by your model.
Similar to the analogous function in Step 5, your function should have three steps:
- Extract the bottleneck features corresponding to the chosen CNN model.
- Supply the bottleneck features as input to the model to return the predicted vector. Note that the argmax of this prediction vector gives the index of the predicted dog breed.
- Use the
dog_namesarray defined in Step 0 of this notebook to return the corresponding breed.
The functions to extract the bottleneck features can be found in extract_bottleneck_features.py, and they have been imported in an earlier code cell. To obtain the bottleneck features corresponding to your chosen CNN architecture, you need to use the function
extract_{network}
where {network}, in the above filename, should be one of VGG19, Resnet50, InceptionV3, or Xception.
In [64]:
"""
from keras.applications.resnet50 import ResNet50
# define ResNet50 model
ResNet50_model = ResNet50(weights='imagenet')
def extract_Resnet50(tensor):
from keras.applications.resnet50 import ResNet50, preprocess_input
return ResNet50(weights='imagenet', include_top=False).predict(preprocess_input(tensor))
"""
"""
def Resnet50_predict_breed(img_path):
# extract bottleneck features for Resnet50 Model
img=preprocess_input(path_to_tensor(img_path))
bottleneck_feature = ResNet50(weights='imagenet', include_top=False).predict(img)
# obtain predicted vector
predicted_vector = ResNet_model.predict(bottleneck_feature)
# return dog breed that is predicted by the model
return dog_names[np.argmax(predicted_vector)]
"""
Out[64]:
"\ndef Resnet50_predict_breed(img_path):\n # extract bottleneck features for Resnet50 Model\n img=preprocess_input(path_to_tensor(img_path))\n bottleneck_feature = ResNet50(weights='imagenet', include_top=False).predict(img) \n \n # obtain predicted vector\n predicted_vector = ResNet_model.predict(bottleneck_feature)\n \n # return dog breed that is predicted by the model\n return dog_names[np.argmax(predicted_vector)]\n\n"
In [31]:
### TODO: Write a function that takes a path to an image as input
### and returns the dog breed that is predicted by the model.
from extract_bottleneck_features import *
def Xception_predict_breed(img_path):
# extract bottleneck features for Resnet50 Model
bottleneck_feature = extract_Xception(path_to_tensor(img_path))
# obtain predicted vector
predicted_vector = Xception_model.predict(bottleneck_feature)
# return dog breed that is predicted by the model
return dog_names[np.argmax(predicted_vector)]
Step 6: Write your Algorithm¶
Write an algorithm that accepts a file path to an image and first determines whether the image contains a human, dog, or neither. Then,
- if a dog is detected in the image, return the predicted breed.
- if a human is detected in the image, return the resembling dog breed.
- if neither is detected in the image, provide output that indicates an error.
You are welcome to write your own functions for detecting humans and dogs in images, but feel free to use the face_detector and dog_detector functions developed above. You are required to use your CNN from Step 5 to predict dog breed.
Some sample output for our algorithm is provided below, but feel free to design your own user experience!
(IMPLEMENTATION) Write your Algorithm¶
In [32]:
### TODO: Write your algorithm.
### Feel free to use as many code cells as needed.
from keras.preprocessing import image
from os import walk
from os import listdir
from os.path import isfile, join
import random
import numpy as np
import cv2
def show_image(path):
img = image.load_img(path, target_size=(224, 224))
img = image.img_to_array(img)
plt.imshow(img/255)
plt.show()
def whos_face_is_this(img_path):
if(dog_detector(img_path)):
print("\n**************************************")
show_image(img_path)
print("hello, Doggy!")
print("Your predicted breed is....")
print(Xception_predict_breed(img_path))
elif(humanface_detector(img_path)):
print("\n**************************************")
show_image(img_path)
print("Hello, Human!")
print("You look like a.... ")
print(Xception_predict_breed(img_path))
else:
print("\n**************************************")
show_image(img_path)
print("**No face detected..ERROR..**")
Step 7: Test Your Algorithm¶
In this section, you will take your new algorithm for a spin! What kind of dog does the algorithm think that you look like? If you have a dog, does it predict your dog's breed accurately? If you have a cat, does it mistakenly think that your cat is a dog?
(IMPLEMENTATION) Test Your Algorithm on Sample Images!¶
Test your algorithm at least six images on your computer. Feel free to use any images you like. Use at least two human and two dog images.
Question 6: Is the output better than you expected :) ? Or worse :( ? Provide at least three possible points of improvement for your algorithm.
Answer:
In [40]:
## TODO: Execute your algorithm from Step 6 on
## at least 6 images on your computer.
## Feel free to use as many code cells as needed.
#load test files from dog-project/dog_images
doggy = np.array(glob("dog_images/*"))
print('No. of files:', len(doggy))
No. of files: 12
In [41]:
for d in doggy:
whos_face_is_this(d)
**************************************
hello, Doggy! Your predicted breed is.... Chesapeake_bay_retriever **************************************
hello, Doggy! Your predicted breed is.... Pembroke_welsh_corgi **************************************
hello, Doggy! Your predicted breed is.... Labrador_retriever **************************************
hello, Doggy! Your predicted breed is.... Golden_retriever **************************************
hello, Doggy! Your predicted breed is.... Alaskan_malamute **************************************
hello, Doggy! Your predicted breed is.... Dalmatian **************************************
hello, Doggy! Your predicted breed is.... American_water_spaniel **************************************
hello, Doggy! Your predicted breed is.... Dalmatian **************************************
hello, Doggy! Your predicted breed is.... Collie **************************************
hello, Doggy! Your predicted breed is.... Irish_water_spaniel **************************************
hello, Doggy! Your predicted breed is.... Golden_retriever **************************************
hello, Doggy! Your predicted breed is.... Icelandic_sheepdog
In [42]:
#load test files from dog-project/Hooman_images
hooman = np.array(glob("Hooman_images/*"))
print('No. of files:', len(hooman))
No. of files: 8
In [43]:
for h in hooman:
whos_face_is_this(h)
**************************************
Hello, Human! You look like a.... Chinese_crested **************************************
**No face detected..ERROR..** **************************************
**No face detected..ERROR..** **************************************
hello, Doggy! Your predicted breed is.... Alaskan_malamute **************************************
Hello, Human! You look like a.... Poodle **************************************
Hello, Human! You look like a.... Belgian_tervuren **************************************
**No face detected..ERROR..** **************************************
hello, Doggy! Your predicted breed is.... Alaskan_malamute
In [65]:
#load test files from dog-project/My_Pic
Mypic = np.array(glob("My_Pic/*"))
print('No. of files:', len(Mypic))
No. of files: 2
In [66]:
for my in Mypic:
whos_face_is_this(my)
**************************************
Hello, Human! You look like a.... Bichon_frise **************************************
Hello, Human! You look like a.... Bichon_frise
Question 6. Is the output better than you expected:) ? Or worse :( ? Provide at least three possible points of improvement for your algorithm.
Answer:
Yes, my output is better than i expected as we can see that-
(American_water_spaniel and Irish_water_spaniel) , (Icelandic_sheepdog,Pembroke_welsh_corgi) looks similar in skin colour but they are predicted 100% correctly, same we can see that retriever dogs like Chesapeake_bay_retriever,Labrador_retriever and Golden_retriever are predicted perfectly .
This model predict dog label correctly in group of same dogs(look Chesapeake_bay_retriever).
This model predict dog label for a dog correcty in human with dog image(look Alaskan_malamute with human).
This model predict human label for fake dog(camera filter image)- and gave the most resemble label to it as we can Chinese_crested dog has long hairstyles like that girl in pic.
This model correctly shows error when no face resembleing with human or dog is detected (look cat and tiger).
This model correctly put same label on my two pics.
Provide at least three possible points of improvement for your algorithm.
Earlier as we used harrclassifier for human face detection there we can use one of Bottleneck Features model we used in dog detector to detect human image.
As we can see 2nd last image of girl that was rotated was not identified by algo due to rotation varience hence Image augumentation is needed here to improve accuracy.
This can also be improved if algo can identify correctly more than one dog breed in group of different dogs(above 7th img from starting ,there it only identify Collie dog breed and not the other one).
Please download your notebook to submit¶
In order to submit, please do the following:
- Download an HTML version of the notebook to your computer using 'File: Download as...'
- Click on the orange Jupyter circle on the top left of the workspace.
- Navigate into the dog-project folder to ensure that you are using the provided dog_images, lfw, and bottleneck_features folders; this means that those folders will not appear in the dog-project folder. If they do appear because you downloaded them, delete them.
- While in the dog-project folder, upload the HTML version of this notebook you just downloaded. The upload button is on the top right.
- Navigate back to the home folder by clicking on the two dots next to the folder icon, and then open up a terminal under the 'new' tab on the top right
- Zip the dog-project folder with the following command in the terminal:
zip -r dog-project.zip dog-project - Download the zip file by clicking on the square next to it and selecting 'download'. This will be the zip file you turn in on the next node after this workspace!
In [48]:
!!jupyter nbconvert *.ipynb
Out[48]:
['[NbConvertApp] Converting notebook dog_app.ipynb to html', '[NbConvertApp] Writing 4686258 bytes to dog_app.html']